Autism RNASeq analysis

Preparation

Load packages and set options

Setup the default knitr options for the R code sections below.

Load the packages used throughout this analysis.

setwd("/share/users/Mike/AutismPaper/RNASeq/R")
suppressMessages({
  library('kableExtra')
  library('goseq')
  library('readr')
  library('stringr')
  library('tibble')
  library('dplyr')
  library('tidyr')
  library('VennDiagram')
  library('tximport')
  library('DESeq2')
  library('regionReport')
  library('ggplot2')
  library('BiocParallel')
  library('RColorBrewer')
  library('colorspace')
  library('dendextend')
  library('gplots')
  library('superheat')
  library('GGally')
  library('corrplot')
  library('pheatmap')
  library('DT')
  library('devtools')
  library('VennDiagram')
  library('ggrepel')
  register(MulticoreParam(detectCores()))
  options(tibble.print_max=100)
})

Import metadata

This section assembles the metadata for the datasets included in this analysis.

meta_int <- read_tsv('../../metadata/RNASeq.internal.tsv') %>%
  mutate(library = 'paired end') %>%
  mutate(readlength = 101) %>%
  mutate(tissue = 'SVZ') %>%
  mutate(path = paste0('/share/users/Mike/AutismPaper/RNASeq/OurData/', 
                           sample, '.refseqnorrna.kallisto.paired/abundance.h5')
  )

meta_ext <- read_tsv('../../metadata/RNASeq.external.tsv') %>%
  mutate(library = 'single end') %>%
  mutate(readlength = 75) %>%
  mutate(condition = 'Control') %>%
  mutate(path = paste0('/share/users/Mike/AutismPaper/RNASeq/external/', 
                           sample, '.refseqnorrna.kallisto.single/abundance.h5')
  )

meta_ext1 <-read_tsv('../../metadata/RNASeq.external2.tsv') %>%
  mutate(path = paste0('/share/users/Mike/AutismPaper/RNASeq/external2/', 
                           sample, '.refseqnorna.kallisto.single/abundance.h5') 
)

metadata <- bind_rows(meta_int, meta_ext, meta_ext1) %>%
  mutate(age = factor(if_else(!is.na(age_wpc), 'Fetal', 
                              if_else(between(age_years,1,19),'Child','Adult')),
                      levels = c('Fetal', 'Child', 'Adult')
                      )
         ) %>%
  mutate(tissue1 = if_else(grepl('SVZ$', tissue), 'SVZ', 
                           if_else(grepl('frontal', tissue), 'FC',
                                   if_else(grepl('temporal', tissue), 'TC',
                                           tissue)
                                   )
                           )
         ) %>%
  mutate(sample = if_else(grepl('OurData', path),
                          paste(sep="_",sub("(\\w).*", "\\1", condition), 
                                paste0(age_years, 'y'),
                                sample), 
                          if_else(grepl('external2',path), 
                                  paste(sep='_', sub("(\\w).*", "\\1", condition), 
                                        tissue1, 
                                        paste0(age_years, 'y'), 
                                        sub(".*(\\d{2})$", "\\1", sample)
                                        ),
                                  paste(sep='_', tissue, 
                                        paste0(age_wpc, 'w'), 
                                        sub(".*(\\d{2})$", "\\1", sample)
                                        )
                                  )
                          )
  ) %>%
  filter(is.na(sex) | sex == 'M')

SVZ tissue analysis

DESeq2 analysis

Select SVZ metadata

Select the metadata corresponding to the SVZ tissue samples:

metadata1 <- metadata %>%
  filter(grepl('SVZ',tissue) & !grepl('H16$', sample)) %>%
  mutate(condition = factor(condition, levels=c('Control', 'Autism'))) %>%
  mutate(grp = factor(paste0(age,condition), levels=c('FetalControl', 'ChildControl', 'ChildAutism', 'AdultControl', 'AdultAutism'))) %>%
  arrange(sample) %>%
  dplyr::select( c('sample', 'condition', 'library', 'readlength', 'tissue', 'age', 'grp', 'path'))
metadata1

Read kallisto output files

Read the files with the kallisto estimated counts for each of the transcripts in RefSeq version 108 (http://www.ncbi.nlm.nih.gov/genome/annotation_euk/Homo_sapiens/108/).

files <- metadata1 %>% dplyr::select('sample', 'path') %>% spread('sample', 'path') %>% as.list() %>% unlist()
txi.kallisto <- tximport(files, type = "kallisto", txOut = TRUE)
metadata1.df <- metadata1 %>%  column_to_rownames(var="sample") %>% as.data.frame()
rownames(txi.kallisto$counts) <- as.character(rownames(txi.kallisto$counts))

Read RefSeq transcript-to-genes data

Read the transcript-to-genes metadata table that associates RefSeq accessions (e.g. NM_xxxxxx) to Entrez Gene IDs, symbols, etc.

ttgfile <- '/share/db/refseq/human/refseq.108.rna.t2g.txt'
ttg <- read_tsv(ttgfile)

Perform Wald test

• Create a DESeq dataset from the imported estimated counts object with a model design based on a factor made by combining the age group and the condition (grp) with a baseline level corresponding to the Control condition and Fetal age group.
• Filter out all the rows that have less than 5 reads on average in the fetal or 10 reads in the non-fetal samples.
• Perform a Wald test on this model.
• List the available contrasts

dds <- DESeqDataSetFromTximport(txi = txi.kallisto, colData = metadata1.df, design = ~grp)
# c <- counts(dds) %>% as.tibble()
# c[['target_id']] <- rownames(counts(dds))
# c <- c %>% mutate(sum1 = rowSums(dplyr::select(., contains("_H"))), 
#                   sum2 = rowSums(dplyr::select(., contains("SVZ"))),
#                   sum3 = rowSums(dplyr::select(., starts_with("A_"))),
#                   sum4 = rowSums(dplyr::select(., starts_with("C_"))))
# n1 <- colnames(c) %>% as.tibble() %>% filter(grepl('_H', value)) %>% summarise(n()) %>% as.numeric()
# n2 <- colnames(c) %>% as.tibble() %>% filter(grepl('SVZ', value)) %>% summarise(n()) %>% as.numeric()
# n3 <- colnames(c) %>% as.tibble() %>% filter(grepl('A_', value)) %>% summarise(n()) %>% as.numeric()
# n4 <- colnames(c) %>% as.tibble() %>% filter(grepl('C_', value)) %>% summarise(n()) %>% as.numeric()
# c <- c %>% filter((sum1 > 10 * n1) & (sum2 > 5 * n2) & (sum3 > 10 * n3) & (sum4 > 10 * n4))
# dds <- dds[c$target_id,]
# rm('c')
ddsWald <- DESeq(dds, test = "Wald", betaPrior = TRUE, parallel = TRUE)
resultsNames(ddsWald)

## [1] "Intercept"       "grpFetalControl" "grpChildControl" "grpChildAutism" 
## [5] "grpAdultControl" "grpAdultAutism"

DESeq results

Extract the results from the model.

resultsChild   <- results(ddsWald, alpha = 0.05, 
                          contrast = c('grp', 'ChildControl', 'FetalControl'))
resultsAdult   <- results(ddsWald, alpha = 0.05,
                          contrast = c('grp', 'AdultControl', 'FetalControl'))
resultsChildA  <- results(ddsWald, alpha = 0.05,
                          contrast = c('grp', 'ChildAutism', 'FetalControl'))
resultsAdultA  <- results(ddsWald, alpha = 0.05,
                          contrast = c('grp', 'AdultAutism', 'FetalControl'))
resultsChildCA <- results(ddsWald, alpha = 0.05,
                          contrast = c('grp','ChildControl','ChildAutism'))
resultsAdultCA <- results(ddsWald, alpha = 0.05,
                          contrast = c('grp','AdultControl','AdultAutism'))


# From the DESeq2 vignette: 
#
# "Regarding multiple test correction, if a user is planning to contrast all pairs of many levels
#  and then selectively reporting the results of only a subset of those pairs, one needs to perform multiple testing across contrasts 
#  as well as genes to control for this additional form of multiple testing. 
#  This can be done by using the p.adjust function across a long vector of p values from all pairs of contrasts, 

#  then re-assigning these adjusted p values to the appropriate results table."
# pChild <- resultsChild %>% as.tibble() %>% rownames_to_column(var='target_id') %>%
#   dplyr::select(c('target_id', 'pvalue')) %>% spread('target_id', 'pvalue') %>%
#   as.list() %>% unlist()
# pAdult <- resultsAdult %>% as.tibble() %>% rownames_to_column(var='target_id') %>%
#   dplyr::select(c('target_id', 'pvalue')) %>% spread('target_id', 'pvalue') %>%
#   as.list() %>% unlist()

signifChild <- resultsChild %>%
  as.tibble(.) %>% rownames_to_column(var='target_id') %>%
  filter(padj<0.05 & abs(log2FoldChange)<15) %>% arrange(padj) %>%
  left_join(ttg, by = 'target_id') %>%
  dplyr::select(c('target_id', 'baseMean', 'log2FoldChange', 'lfcSE', 'stat', 'pvalue', 'padj', 'gene_id', 'symbol', 'description', 'status', 'molecule_type'))

signifAdult <- resultsAdult %>%
  as.tibble(.) %>% rownames_to_column(var='target_id') %>%
  filter(padj<0.05 & abs(log2FoldChange)<15) %>% arrange(padj) %>%
  left_join(ttg, by = 'target_id') %>%
  dplyr::select(c('target_id', 'baseMean', 'log2FoldChange', 'lfcSE', 'stat', 'pvalue', 'padj', 'gene_id', 'symbol', 'description', 'status', 'molecule_type'))

signifChildA <- resultsChildA %>%
  as.tibble(.) %>% rownames_to_column(var='target_id') %>%
  filter(padj<0.05 & abs(log2FoldChange)<15) %>% arrange(padj) %>%
  left_join(ttg, by = 'target_id') %>%
  dplyr::select(c('target_id', 'baseMean', 'log2FoldChange', 'lfcSE', 'stat', 'pvalue', 'padj', 'gene_id', 'symbol', 'description', 'status', 'molecule_type'))

signifAdultA <- resultsAdultA %>%
  as.tibble(.) %>% rownames_to_column(var='target_id') %>%
  filter(padj<0.05 & abs(log2FoldChange)<15) %>% arrange(padj) %>%
  left_join(ttg, by = 'target_id') %>%
  dplyr::select(c('target_id', 'baseMean', 'log2FoldChange', 'lfcSE', 'stat', 'pvalue', 'padj', 'gene_id', 'symbol', 'description', 'status', 'molecule_type'))

signifChildCA <- resultsChildCA %>%
  as.tibble(.) %>% rownames_to_column(var='target_id') %>%
  filter(padj<0.05 & abs(log2FoldChange)<15) %>% arrange(padj) %>%
  left_join(ttg, by = 'target_id') %>%
  dplyr::select(c('target_id', 'baseMean', 'log2FoldChange', 'lfcSE', 'stat', 'pvalue', 'padj', 'gene_id', 'symbol', 'description', 'status', 'molecule_type'))

signifAdultCA <- resultsAdultCA %>%
  as.tibble(.) %>% rownames_to_column(var='target_id') %>%
  filter(padj<0.05 & abs(log2FoldChange)<15) %>% arrange(padj) %>%
  left_join(ttg, by = 'target_id') %>%
  dplyr::select(c('target_id', 'baseMean', 'log2FoldChange', 'lfcSE', 'stat', 'pvalue', 'padj', 'gene_id', 'symbol', 'description', 'status', 'molecule_type'))

devChildCA  <- intersect(signifChild$target_id, signifChildCA$target_id)
devAdultCA  <- intersect(signifAdult$target_id, signifAdultCA$target_id)
devChildCAg <- tibble(target_id=devChildCA) %>% left_join(ttg, by='target_id') %>% 
  dplyr::select('gene_id') %>% as.list() %>% unlist() %>% unique()
devAdultCAg <- tibble(target_id=devAdultCA) %>% left_join(ttg, by='target_id') %>% 
  dplyr::select('gene_id') %>% as.list() %>% unlist() %>% unique()
devChildCAs <- tibble(target_id=devChildCA) %>% left_join(ttg, by='target_id') %>% 
  dplyr::select('symbol') %>% as.list() %>% unlist() %>% unique()
devAdultCAs <- tibble(target_id=devAdultCA) %>% left_join(ttg, by='target_id') %>% 
  dplyr::select('symbol') %>% as.list() %>% unlist() %>% unique()

Control - Fetal

Summary of the number of transcripts and genes differentially expressed in control samples of different age groups vs. the fetal samples (“developmental genes”):

rsum_control_fetal <- tribble( ~Counts, ~Child, ~Adult,
                               'Transcripts', 
                               length(signifChild$target_id), 
                               length(signifAdult$target_id), 
                               'Genes', 
                               length(unique(signifChild$gene_id)),
                               length(unique(signifAdult$gene_id))
                               )
kable(rsum_control_fetal, format = 'markdown')

Counts	Child	Adult
Transcripts	15767	18111
Genes	10248	11265

Venn diagram:

grid.draw(venn.diagram(list(Child=signifChild$target_id, Adult=signifAdult$target_id), 
             filename = NULL,
             alpha = 0.5,
             inverted = TRUE, 
             cat.pos = c(45, 315),
             fill = c("cornflowerblue", "darkorchid1"),
             cat.col = c("darkblue", "darkorchid4")
             )
)

Autism - Fetal

Summary of the number of transcripts and genes differentially expressed in autism samples of different age groups vs. the fetal samples:

rsum_autism_fetal <- tribble( ~Counts, ~Child, ~Adult, 
                               'Transcripts', 
                               length(signifChildA$target_id), 
                               length(signifAdultA$target_id), 
                               'Genes', 
                               length(unique(signifChildA$gene_id)),
                               length(unique(signifAdultA$gene_id))
                               )
kable(rsum_autism_fetal, format = 'markdown')

Counts	Child	Adult
Transcripts	18621	17554
Genes	11626	11089

Venn diagram:

grid.draw(venn.diagram(list(Child=signifChildA$target_id, Adult=signifAdultA$target_id),
             filename = NULL,
             alpha = 0.5,
             fill = c("cornflowerblue", "darkorchid1"),
             cat.col = c("darkblue", "darkorchid4")
             )
)

Control - Autism

Summary of the number of transcripts and genes differentially expressed in autism samples of different age groups vs. the same age control samples:

rsum_autism_control <- tribble( ~Counts, ~Child, ~Adult,
                               'Transcripts', 
                               length(signifChildCA$target_id), 
                               length(signifAdultCA$target_id), 
                               'Genes', 
                               length(unique(signifChildCA$gene_id)),
                               length(unique(signifAdultCA$gene_id))
                               )
kable(rsum_autism_control, format = 'markdown')

Counts	Child	Adult
Transcripts	32	54
Genes	32	50

Venn diagram:

grid.draw(venn.diagram(list(Child=signifChildCA$target_id, Adult=signifAdultCA$target_id),
                       inverted = TRUE, 
                       cat.pos = c(45, 315),
                       filename = NULL,
                       alpha = 0.5,
                       fill = c("cornflowerblue", "darkorchid1"),
                       cat.col = c("darkblue", "darkorchid4")
                       )
          )

Developmental Control - Autism

Summary of the number of transcripts and genes differentially expressed in autism samples of different age groups vs. the same age control samples AND also different in the control age group vs. the fetal samples (“developmental”):

rsum_autism_control_fetal <- tribble( ~Counts, ~Child, ~Adult,
                               'Transcripts', 
                               length(devChildCA), 
                               length(devAdultCA), 
                               'Genes', 
                               length(devChildCAg),
                               length(devAdultCAg)
                               )
kable(rsum_autism_control_fetal, format = 'markdown')

Counts	Child	Adult
Transcripts	22	44
Genes	22	40

Venn diagram:

grid.draw(venn.diagram(list(Child=devChildCA, Adult=devAdultCA),
                       filename = NULL,
                       alpha = 0.5,
                       fill = c("cornflowerblue", "darkorchid1"),
                       cat.col = c("darkblue", "darkorchid4")
                       )
          )

QC

PCA

## Transform count data
intgroup = "grp"
rld <- tryCatch(rlog(dds), error = function(e) { rlog(dds, fitType = 'mean') })

## Perform PCA analysis and get percent of variance explained
data_pca <- plotPCA(rld, intgroup = intgroup, ntop = 1000, returnData = TRUE)
percentVar <- round(100 * attr(data_pca, "percentVar"))

## Make plot
data_pca %>%
ggplot(aes_string(x = "PC1", y = "PC2", color = "grp")) +
 geom_point(size = 3) +
 xlab(paste0("PC1: ", percentVar[1], "% variance")) +
 ylab(paste0("PC2: ", percentVar[2], "% variance")) +
 coord_fixed() +
 geom_text(aes(label=name), nudge_x = 1, hjust = 0, size = 3, check_overlap = FALSE)

The above plot shows the first two principal components that explain the variability in the data using the regularized log count data. If you are unfamiliar with principal component analysis, you might want to check the Wikipedia entry or this interactive explanation. In this case, the first and second principal component explain 12 and 7 percent of the variance respectively.

Sample-to-sample distances

## Obtain the sample euclidean distances
sampleDists <- dist(t(assay(rld)))
sampleDistMatrix <- as.matrix(sampleDists)
## Add names based on intgroup
rownames(sampleDistMatrix) <- rownames(colData(rld))
colnames(sampleDistMatrix) <- NULL

## Define colors to use for the heatmap if none were supplied
colors <- colorRampPalette( rev(brewer.pal(9, "Blues")) )(255)

## Make the heatmap
pheatmap(sampleDistMatrix, clustering_distance_rows = sampleDists, 
         clustering_distance_cols = sampleDists, color = colors)

This plot shows how samples are clustered based on their euclidean distance using the regularized log transformed count data. This figure gives an overview of how the samples are hierarchically clustered. It is a complementary figure to the PCA plot.

MA Plots

Control - Fetal

This section contains the MA plots (see Wikipedia) that compare the mean of the normalized counts against the log fold change. They show one point per feature. The points are shown in red if the feature has an adjusted p-value less than alpha, that is, the statistically significant features are shown in red.

ChildControl-Fetal

## MA plot with alpha used in DESeq2::results()
plotMA(resultsChild, alpha = 0.05, main = 'MA plot with alpha = 0.05')
abline(h=c(-1,1),col="dodgerblue",lwd=2)

This plot uses alpha = 0.05, which is the alpha value used to determine which resulting features were significant when running the function DESeq2::results().

AdultControl-Fetal

## MA plot with alpha used in DESeq2::results()
plotMA(resultsAdult, alpha = 0.05, main = 'MA plot with alpha = 0.05')
abline(h=c(-1,1),col="dodgerblue",lwd=2)

This plot uses alpha = 0.05, which is the alpha value used to determine which resulting features were significant when running the function DESeq2::results().

Autism - Fetal

This section contains the MA plots (see Wikipedia) that compare the mean of the normalized counts against the log fold change. They show one point per feature. The points are shown in red if the feature has an adjusted p-value less than alpha, that is, the statistically significant features are shown in red.

ChildAutism-Fetal

## MA plot with alpha used in DESeq2::results()
plotMA(resultsChildA, alpha = 0.05, main = 'MA plot with alpha = 0.05')
abline(h=c(-1,1),col="dodgerblue",lwd=2)

This plot uses alpha = 0.05, which is the alpha value used to determine which resulting features were significant when running the function DESeq2::results().

AdultAutism-Fetal

## MA plot with alpha used in DESeq2::results()
plotMA(resultsAdultA, alpha = 0.05, main = 'MA plot with alpha = 0.05')
abline(h=c(-1,1),col="dodgerblue",lwd=2)

This plot uses alpha = 0.05, which is the alpha value used to determine which resulting features were significant when running the function DESeq2::results().

Control - Autism

This section contains the MA plots (see Wikipedia) that compare the mean of the normalized counts against the log fold change. They show one point per feature. The points are shown in red if the feature has an adjusted p-value less than alpha, that is, the statistically significant features are shown in red.

Child

## MA plot with alpha used in DESeq2::results()
plotMA(resultsChildCA, alpha = 0.05, main = 'MA plot with alpha = 0.05')
abline(h=c(-1,1),col="dodgerblue",lwd=2)

This plot uses alpha = 0.05, which is the alpha value used to determine which resulting features were significant when running the function DESeq2::results().

Adult

## MA plot with alpha used in DESeq2::results()
plotMA(resultsAdultCA, alpha = 0.05, main = 'MA plot with alpha = 0.05')
abline(h=c(-1,1),col="dodgerblue",lwd=2)

This plot uses alpha = 0.05, which is the alpha value used to determine which resulting features were significant when running the function DESeq2::results().

P-values distributions

Control - Fetal

This plot shows a histogram of the unadjusted p-values. It might be skewed right or left, or flat as shown in the Wikipedia examples. The shape depends on the percent of features that are differentially expressed. For further information on how to interpret a histogram of p-values check David Robinson’s post on this topic.

ChildControl-Fetal

## P-value histogram plot
resultsChild %>% as.tibble() %>%
  filter(!is.na(pvalue)) %>%
  ggplot(aes(x = pvalue)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title='Histogram of unadjusted p-values') +
  xlab('Unadjusted p-values') +
  coord_cartesian(xlim=c(0, 1.0005), ylim=c(0,20000))

This is the numerical summary of the distribution of the p-values.

## P-value distribution summary
res <- resultsChild %>%
  as.tibble() %>%
  filter(!is.na(pvalue))
summary(res$pvalue)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  0.0000  0.1305  0.5022  0.4522  0.7089  1.0000

AdultControl-Fetal

## P-value histogram plot
resultsAdult %>% as.tibble() %>%
  filter(!is.na(pvalue)) %>%
  ggplot(aes(x = pvalue)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title='Histogram of unadjusted p-values') +
  xlab('Unadjusted p-values') +
  coord_cartesian(xlim=c(0, 1.0005), ylim=c(0,20000))

This is the numerical summary of the distribution of the p-values.

## P-value distribution summary
res <- resultsAdult %>%
  as.tibble() %>%
  filter(!is.na(pvalue))
summary(res$pvalue)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  0.0000  0.1024  0.4813  0.4386  0.7030  1.0000

Autism - Fetal

This plot shows a histogram of the unadjusted p-values. It might be skewed right or left, or flat as shown in the Wikipedia examples. The shape depends on the percent of features that are differentially expressed. For further information on how to interpret a histogram of p-values check David Robinson’s post on this topic.

ChildAutism-Fetal

## P-value histogram plot
resultsChildA %>% as.tibble() %>%
  filter(!is.na(pvalue)) %>%
  ggplot(aes(x = pvalue)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title='Histogram of unadjusted p-values') +
  xlab('Unadjusted p-values') +
  coord_cartesian(xlim=c(0, 1.0005), ylim=c(0,20000))

This is the numerical summary of the distribution of the p-values.

## P-value distribution summary
res <- resultsChildA %>%
  as.tibble() %>%
  filter(!is.na(pvalue))
summary(res$pvalue)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  0.0000  0.1083  0.4934  0.4415  0.6958  1.0000

AdultAutism-Fetal

## P-value histogram plot
resultsAdultA %>% as.tibble() %>%
  filter(!is.na(pvalue)) %>%
  ggplot(aes(x = pvalue)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title='Histogram of unadjusted p-values') +
  xlab('Unadjusted p-values') +
  coord_cartesian(xlim=c(0, 1.0005), ylim=c(0,20000))

This is the numerical summary of the distribution of the p-values.

## P-value distribution summary
res <- resultsAdultA %>%
  as.tibble() %>%
  filter(!is.na(pvalue))
summary(res$pvalue)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  0.0000  0.1054  0.4815  0.4398  0.7025  1.0000

Control - Autism

This plot shows a histogram of the unadjusted p-values. It might be skewed right or left, or flat as shown in the Wikipedia examples. The shape depends on the percent of features that are differentially expressed. For further information on how to interpret a histogram of p-values check David Robinson’s post on this topic.

Child

## P-value histogram plot
resultsChildCA %>% as.tibble() %>%
  filter(!is.na(pvalue)) %>%
  ggplot(aes(x = pvalue)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title='Histogram of unadjusted p-values') +
  xlab('Unadjusted p-values') +
  coord_cartesian(xlim=c(0, 1.0005), ylim=c(0,20000))

This is the numerical summary of the distribution of the p-values.

## P-value distribution summary
res <- resultsChildCA %>%
  as.tibble() %>%
  filter(!is.na(pvalue))
summary(res$pvalue)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  0.0000  0.5233  0.7569  0.6924  0.9113  1.0000

Adult

## P-value histogram plot
resultsAdultCA %>% as.tibble() %>%
  filter(!is.na(pvalue)) %>%
  ggplot(aes(x = pvalue)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title='Histogram of unadjusted p-values') +
  xlab('Unadjusted p-values') +
  coord_cartesian(xlim=c(0, 1.0005), ylim=c(0,20000))

This is the numerical summary of the distribution of the p-values.

## P-value distribution summary
res <- resultsAdultCA %>%
  as.tibble() %>%
  filter(!is.na(pvalue))
summary(res$pvalue)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  0.0000  0.5294  0.7607  0.6997  0.9210  1.0000

P-values cumulative sums

This table shows the number of features with p-values less or equal than some commonly used cutoff values.

Control - Fetal

ChildControl-Fetal

## Split features by different p-value cutoffs
pval_table <- resultsChild %>% 
  as.tibble() %>% 
  filter(!is.na(pvalue)) %>%
  mutate(Cut = as.numeric(
    as.character(
      cut(pvalue, include.lowest = TRUE, 
          breaks=c(0, 1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1),
          labels = c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1))))) %>%
  group_by(Cut) %>%
  summarise(nCut=n()) %>%
  mutate(Count=cumsum(nCut)) %>%
  dplyr::select(c('Cut', 'Count'))
  
# pval_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
#     0.6, 0.7, 0.8, 0.9, 1), function(x) {
#     data.frame('Cut' = x, 'Count' = sum(resultsChild$pvalue <= x, na.rm = TRUE))
# })
# pval_table <- do.call(rbind, pval_table)
if (outputIsHTML) {
    kable(pval_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(pval_table)
}

Cut	Count
0.0001	7481
0.0010	10365
0.0100	15921
0.0250	19791
0.0500	24011
0.1000	30702
0.2000	40631
0.3000	49345
0.4000	58044
0.5000	67450
0.6000	84321
0.7000	100527
0.8000	112796
0.9000	124479
1.0000	135640

AdultControl-Fetal

## Split features by different p-value cutoffs
pval_table <- resultsAdult %>% 
  as.tibble() %>% 
  filter(!is.na(pvalue)) %>%
  mutate(Cut = as.numeric(
    as.character(
      cut(pvalue, include.lowest = TRUE, 
          breaks=c(0, 1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1),
          labels = c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1))))) %>%
  group_by(Cut) %>%
  summarise(nCut=n()) %>%
  mutate(Count=cumsum(nCut)) %>%
  dplyr::select(c('Cut', 'Count'))
  
# pval_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
#     0.6, 0.7, 0.8, 0.9, 1), function(x) {
#     data.frame('Cut' = x, 'Count' = sum(resultsAdult$pvalue <= x, na.rm = TRUE))
# })
# pval_table <- do.call(rbind, pval_table)
if(outputIsHTML) {
    kable(pval_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(pval_table)
}

Cut	Count
0.0001	8447
0.0010	11763
0.0100	17926
0.0250	22142
0.0500	26746
0.1000	33640
0.2000	43757
0.3000	52507
0.4000	60773
0.5000	69608
0.6000	84381
0.7000	101200
0.8000	113968
0.9000	125172
1.0000	135640

Autism - Fetal

ChildAutism-Fetal

## Split features by different p-value cutoffs
pval_table <- resultsChildA %>% 
  as.tibble() %>% 
  filter(!is.na(pvalue)) %>%
  mutate(Cut = as.numeric(
    as.character(
      cut(pvalue, include.lowest = TRUE, 
          breaks=c(0, 1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1),
          labels = c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1))))) %>%
  group_by(Cut) %>%
  summarise(nCut=n()) %>%
  mutate(Count=cumsum(nCut)) %>%
  dplyr::select(c('Cut', 'Count'))
  
# pval_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
#     0.6, 0.7, 0.8, 0.9, 1), function(x) {
#     data.frame('Cut' = x, 'Count' = sum(resultsChild$pvalue <= x, na.rm = TRUE))
# })
# pval_table <- do.call(rbind, pval_table)
if(outputIsHTML) {
    kable(pval_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(pval_table)
}

Cut	Count
0.0001	9004
0.0010	12251
0.0100	18180
0.0250	22312
0.0500	26710
0.1000	32980
0.2000	42850
0.3000	51276
0.4000	59454
0.5000	68513
0.6000	82434
0.7000	102391
0.8000	114580
0.9000	125681
1.0000	135640

AdultAutism-Fetal

## Split features by different p-value cutoffs
pval_table <- resultsAdultA %>% 
  as.tibble() %>% 
  filter(!is.na(pvalue)) %>%
  mutate(Cut = as.numeric(
    as.character(
      cut(pvalue, include.lowest = TRUE, 
          breaks=c(0, 1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1),
          labels = c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1))))) %>%
  group_by(Cut) %>%
  summarise(nCut=n()) %>%
  mutate(Count=cumsum(nCut)) %>%
  dplyr::select(c('Cut', 'Count'))
  
# pval_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
#     0.6, 0.7, 0.8, 0.9, 1), function(x) {
#     data.frame('Cut' = x, 'Count' = sum(resultsAdult$pvalue <= x, na.rm = TRUE))
# })
# pval_table <- do.call(rbind, pval_table)
if(outputIsHTML) {
    kable(pval_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(pval_table)
}

Cut	Count
0.0001	8092
0.0010	11364
0.0100	17473
0.0250	21766
0.0500	26356
0.1000	33274
0.2000	43444
0.3000	52222
0.4000	60630
0.5000	69683
0.6000	84162
0.7000	101379
0.8000	113652
0.9000	125001
1.0000	135640

Control - Autism

Child

## Split features by different p-value cutoffs
pval_table <- resultsChildCA %>% 
  as.tibble() %>% 
  filter(!is.na(pvalue)) %>%
  mutate(Cut = as.numeric(
    as.character(
      cut(pvalue, include.lowest = TRUE, 
          breaks=c(0, 1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1),
          labels = c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1))))) %>%
  group_by(Cut) %>%
  summarise(nCut=n()) %>%
  mutate(Count=cumsum(nCut)) %>%
  dplyr::select(c('Cut', 'Count'))
  
# pval_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
#     0.6, 0.7, 0.8, 0.9, 1), function(x) {
#     data.frame('Cut' = x, 'Count' = sum(resultsChild$pvalue <= x, na.rm = TRUE))
# })
# pval_table <- do.call(rbind, pval_table)
if(outputIsHTML) {
    kable(pval_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(pval_table)
}

Cut	Count
0.0001	52
0.0010	172
0.0100	623
0.0250	1211
0.0500	2055
0.1000	4029
0.2000	8077
0.3000	13371
0.4000	21268
0.5000	31255
0.6000	43498
0.7000	58250
0.8000	76015
0.9000	98196
1.0000	135640

Adult

## Split features by different p-value cutoffs
pval_table <- resultsAdultCA %>% 
  as.tibble() %>% 
  filter(!is.na(pvalue)) %>%
  mutate(Cut = as.numeric(
    as.character(
      cut(pvalue, include.lowest = TRUE, 
          breaks=c(0, 1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1),
          labels = c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1))))) %>%
  group_by(Cut) %>%
  summarise(nCut=n()) %>%
  mutate(Count=cumsum(nCut)) %>%
  dplyr::select(c('Cut', 'Count'))
  
# pval_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
#     0.6, 0.7, 0.8, 0.9, 1), function(x) {
#     data.frame('Cut' = x, 'Count' = sum(resultsAdult$pvalue <= x, na.rm = TRUE))
# })
# pval_table <- do.call(rbind, pval_table)
if(outputIsHTML) {
    kable(pval_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(pval_table)
}

Cut	Count
0.0001	75
0.0010	206
0.0100	707
0.0250	1263
0.0500	2106
0.1000	4364
0.2000	8379
0.3000	13386
0.4000	20933
0.5000	30670
0.6000	42563
0.7000	57138
0.8000	75165
0.9000	97843
1.0000	135640

Adjusted p-values distributions

Control - Fetal

ChildControl-Fetal

This plot shows a histogram of the BH adjusted p-values. It might be skewed right or left, or flat as shown in the Wikipedia examples.

## Adjusted p-values histogram plot
hdescription <- elementMetadata(resultsChild) %>%
  as.tibble() %>%
  filter(grepl('adjusted',description)) %>%
  dplyr::select('description') %>%
  as.character()
resultsChild %>% 
  as.tibble() %>%
  filter(!is.na(padj)) %>% 
  ggplot(aes(x = padj)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title=paste('Histogram of',hdescription)) +
  xlab('Adjusted p-values') +
  xlim(c(0, 1.0005))

This is the numerical summary of the distribution of the BH adjusted p-values.

## Adjusted p-values distribution summary
res <- resultsChild %>%
  as.tibble() %>%
  filter(!is.na(padj))
summary(res$padj)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  0.0000  0.1128  0.4641  0.4616  0.7879  1.0000

AdultControl-Fetal

This plot shows a histogram of the BH adjusted p-values. It might be skewed right or left, or flat as shown in the Wikipedia examples.

## Adjusted p-values histogram plot
hdescription <- elementMetadata(resultsAdult) %>%
  as.tibble() %>%
  filter(grepl('adjusted',description)) %>%
  dplyr::select('description') %>%
  as.character()
resultsAdult %>% 
  as.tibble() %>%
  filter(!is.na(padj)) %>% 
  ggplot(aes(x = padj)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title=paste('Histogram of',hdescription)) +
  xlab('Adjusted p-values') +
  xlim(c(0, 1.0005))

This is the numerical summary of the distribution of the BH adjusted p-values.

## Adjusted p-values distribution summary
res <- resultsAdult %>%
  as.tibble() %>%
  filter(!is.na(padj))
summary(res$padj)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
## 0.00000 0.08752 0.42321 0.43837 0.75748 1.00000

Autism - Fetal

ChildAutism-Fetal

This plot shows a histogram of the BH adjusted p-values. It might be skewed right or left, or flat as shown in the Wikipedia examples.

## Adjusted p-values histogram plot
hdescription <- elementMetadata(resultsChildA) %>%
  as.tibble() %>%
  filter(grepl('adjusted',description)) %>%
  dplyr::select('description') %>%
  as.character()
resultsChildA %>% 
  as.tibble() %>%
  filter(!is.na(padj)) %>% 
  ggplot(aes(x = padj)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title=paste('Histogram of',hdescription)) +
  xlab('Adjusted p-values') +
  xlim(c(0, 1.0005))

This is the numerical summary of the distribution of the BH adjusted p-values.

## Adjusted p-values distribution summary
res <- resultsChildA %>%
  as.tibble() %>%
  filter(!is.na(padj))
summary(res$padj)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  0.0000  0.0670  0.3939  0.4240  0.7477  1.0000

AdultAutism-Fetal

This plot shows a histogram of the BH adjusted p-values. It might be skewed right or left, or flat as shown in the Wikipedia examples.

## Adjusted p-values histogram plot
hdescription <- elementMetadata(resultsAdultA) %>%
  as.tibble() %>%
  filter(grepl('adjusted',description)) %>%
  dplyr::select('description') %>%
  as.character()
resultsAdultA %>% 
  as.tibble() %>%
  filter(!is.na(padj)) %>% 
  ggplot(aes(x = padj)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title=paste('Histogram of',hdescription)) +
  xlab('Adjusted p-values') +
  xlim(c(0, 1.0005))

This is the numerical summary of the distribution of the BH adjusted p-values.

## Adjusted p-values distribution summary
res <- resultsAdultA %>%
  as.tibble() %>%
  filter(!is.na(padj))
summary(res$padj)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
## 0.00000 0.09385 0.42864 0.44133 0.75822 0.99999

Control - Autism

Child

This plot shows a histogram of the BH adjusted p-values. It might be skewed right or left, or flat as shown in the Wikipedia examples.

## Adjusted p-values histogram plot
hdescription <- elementMetadata(resultsChildCA) %>%
  as.tibble() %>%
  filter(grepl('adjusted',description)) %>%
  dplyr::select('description') %>%
  as.character()
resultsChildCA %>% 
  as.tibble() %>%
  filter(!is.na(padj)) %>% 
  ggplot(aes(x = padj)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title=paste('Histogram of',hdescription)) +
  xlab('Adjusted p-values') +
  coord_cartesian(xlim=c(0, 1.0005), ylim=c(0,1000))

This is the numerical summary of the distribution of the BH adjusted p-values.

## Adjusted p-values distribution summary
res <- resultsChildCA %>%
  as.tibble() %>%
  filter(!is.na(padj))
summary(res$padj)

##     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
## 0.002843 0.999994 0.999994 0.997075 0.999994 0.999997

Adult

This plot shows a histogram of the BH adjusted p-values. It might be skewed right or left, or flat as shown in the Wikipedia examples.

## Adjusted p-values histogram plot
hdescription <- elementMetadata(resultsAdultCA) %>%
  as.tibble() %>%
  filter(grepl('adjusted',description)) %>%
  dplyr::select('description') %>%
  as.character()
resultsAdultCA %>% 
  as.tibble() %>%
  filter(!is.na(padj)) %>% 
  ggplot(aes(x = padj)) +
  geom_histogram(alpha=.5, position='identity', bins = 50) +
  labs(title=paste('Histogram of',hdescription)) +
  xlab('Adjusted p-values') +
  coord_cartesian(xlim=c(0, 1.0005), ylim=c(0,1000))

This is the numerical summary of the distribution of the BH adjusted p-values.

## Adjusted p-values distribution summary
res <- resultsAdultCA %>%
  as.tibble() %>%
  filter(!is.na(padj))
summary(res$padj)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  0.0000  1.0000  1.0000  0.9964  1.0000  1.0000

Adjusted p-values cummulative sums

Control - Fetal

ChildControl-Fetal

This table shows the number of features with BH adjusted p-values less or equal than some commonly used cutoff values.

## Split features by different adjusted p-value cutoffs
padj_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
    0.6, 0.7, 0.8, 0.9, 1), function(x) {
    data.frame('Cut' = x, 'Count' = sum(resultsChild$padj <= x, na.rm = TRUE))
})
padj_table <- do.call(rbind, padj_table)
if(outputIsHTML) {
    kable(padj_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(padj_table)
}

Cut	Count
0.0001	5410
0.0010	7399
0.0100	10862
0.0250	13269
0.0500	15771
0.1000	19429
0.2000	25279
0.3000	31087
0.4000	36774
0.5000	42250
0.6000	48120
0.7000	54204
0.8000	61362
0.9000	69964
1.0000	80578

AdultControl-Fetal

This table shows the number of features with BH adjusted p-values less or equal than some commonly used cutoff values.

## Split features by different adjusted p-value cutoffs
padj_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
    0.6, 0.7, 0.8, 0.9, 1), function(x) {
    data.frame('Cut' = x, 'Count' = sum(resultsAdult$padj <= x, na.rm = TRUE))
})
padj_table <- do.call(rbind, padj_table)
if(outputIsHTML) {
    kable(padj_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(padj_table)
}

Cut	Count
0.0001	6291
0.0010	8440
0.0100	12557
0.0250	15253
0.0500	18114
0.1000	22228
0.2000	28785
0.3000	34956
0.4000	41211
0.5000	47279
0.6000	53584
0.7000	59889
0.8000	67217
0.9000	75578
1.0000	85353

Autism - Fetal

ChildAutism-Fetal

This table shows the number of features with BH adjusted p-values less or equal than some commonly used cutoff values.

## Split features by different adjusted p-value cutoffs
padj_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
    0.6, 0.7, 0.8, 0.9, 1), function(x) {
    data.frame('Cut' = x, 'Count' = sum(resultsChildA$padj <= x, na.rm = TRUE))
})
padj_table <- do.call(rbind, padj_table)
if(outputIsHTML) {
    kable(padj_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(padj_table)
}

Cut	Count
0.0001	6871
0.0010	9158
0.0100	13218
0.0250	15836
0.0500	18625
0.1000	22739
0.2000	29085
0.3000	34914
0.4000	40576
0.5000	45947
0.6000	51369
0.7000	57443
0.8000	63988
0.9000	71534
1.0000	80578

AdultAutism-Fetal

This table shows the number of features with BH adjusted p-values less or equal than some commonly used cutoff values.

## Split features by different adjusted p-value cutoffs
padj_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
    0.6, 0.7, 0.8, 0.9, 1), function(x) {
    data.frame('Cut' = x, 'Count' = sum(resultsAdultA$padj <= x, na.rm = TRUE))
})
padj_table <- do.call(rbind, padj_table)
if(outputIsHTML) {
    kable(padj_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(padj_table)
}

Cut	Count
0.0001	5907
0.0010	8020
0.0100	12033
0.0250	14625
0.0500	17555
0.1000	21789
0.2000	28340
0.3000	34531
0.4000	41018
0.5000	47124
0.6000	53407
0.7000	59945
0.8000	67171
0.9000	75321
1.0000	85353

Control - Autism

Child

This table shows the number of features with BH adjusted p-values less or equal than some commonly used cutoff values.

## Split features by different adjusted p-value cutoffs
padj_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
    0.6, 0.7, 0.8, 0.9, 1), function(x) {
    data.frame('Cut' = x, 'Count' = sum(resultsChildCA$padj <= x, na.rm = TRUE))
})
padj_table <- do.call(rbind, padj_table)
if(outputIsHTML) {
    kable(padj_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(padj_table)
}

Cut	Count
0.0001	0
0.0010	0
0.0100	10
0.0250	23
0.0500	32
0.1000	40
0.2000	86
0.3000	105
0.4000	166
0.5000	182
0.6000	229
0.7000	290
0.8000	326
0.9000	388
1.0000	71079

Adult

This table shows the number of features with BH adjusted p-values less or equal than some commonly used cutoff values.

## Split features by different adjusted p-value cutoffs
padj_table <- lapply(c(1e-04, 0.001, 0.01, 0.025, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5,
    0.6, 0.7, 0.8, 0.9, 1), function(x) {
    data.frame('Cut' = x, 'Count' = sum(resultsAdultCA$padj <= x, na.rm = TRUE))
})
padj_table <- do.call(rbind, padj_table)
if(outputIsHTML) {
    kable(padj_table, format = 'markdown', align = c('c', 'c'))
} else {
    kable(padj_table)
}

Cut	Count
0.0001	3
0.0010	3
0.0100	17
0.0250	32
0.0500	54
0.1000	75
0.2000	105
0.3000	145
0.4000	210
0.5000	260
0.6000	320
0.7000	376
0.8000	464
0.9000	524
1.0000	78193

Volcano plots

Control - Fetal

ChildControl - Fetal

resultsChild %>%
  as.tibble()  %>%
  rownames_to_column(var='target_id') %>%
  mutate(sig=ifelse(target_id %in% signifChild[['target_id']], "FDR<0.05", "Not Sig")) %>%
  left_join(ttg, by = 'target_id') %>% 
  ggplot(aes(log2FoldChange, -log10(padj))) + 
  geom_point(aes(col=sig)) + 
  scale_color_manual(values=c("red", "black"))

#  coord_cartesian(ylim=c(0,100))

AdultControl - Fetal

resultsAdult %>%
  as.tibble()  %>%
  rownames_to_column(var='target_id') %>%
  mutate(sig=ifelse(target_id %in% signifAdult[['target_id']], "FDR<0.05", "Not Sig")) %>%
  left_join(ttg, by = 'target_id') %>% 
  ggplot(aes(log2FoldChange, -log10(padj))) + 
  geom_point(aes(col=sig)) + 
  scale_color_manual(values=c("red", "black"))

  #coord_cartesian(ylim=c(0,100))

Autism - Fetal

ChildAutism - Fetal

resultsChildA %>%
  as.tibble()  %>%
  rownames_to_column(var='target_id') %>%
  mutate(sig=ifelse(target_id %in% signifChildA[['target_id']], "FDR<0.05", "Not Sig")) %>%
  left_join(ttg, by = 'target_id') %>% 
  ggplot(aes(log2FoldChange, -log10(padj))) + 
  geom_point(aes(col=sig)) + 
  scale_color_manual(values=c("red", "black"))

#  coord_cartesian(ylim=c(0,100))

AdultAutism - Fetal

resultsAdultA %>%
  as.tibble()  %>%
  rownames_to_column(var='target_id') %>%
  mutate(sig=ifelse(target_id %in% signifAdultA[['target_id']], "FDR<0.05", "Not Sig")) %>%
  left_join(ttg, by = 'target_id') %>% 
  ggplot(aes(log2FoldChange, -log10(padj))) + 
  geom_point(aes(col=sig)) + 
  scale_color_manual(values=c("red", "black"))

#  coord_cartesian(ylim=c(0,100))

Control - Autism

Child

r1 <- resultsChildCA %>%
  as.tibble()  %>%
  rownames_to_column(var='target_id') %>%
  mutate(sig=ifelse(target_id %in% signifChildCA[['target_id']], "FDR<0.05", "Not Sig")) %>%
  left_join(ttg, by = 'target_id')
r1 %>% ggplot(aes(log2FoldChange, -log10(padj))) + 
  geom_point(aes(col=sig)) + 
  scale_color_manual(values=c("red", "black")) +
#  coord_cartesian(ylim=c(0,10)) + 
  geom_text_repel(data=filter(r1, sig=='FDR<0.05'), aes(label=symbol))

Adult

r1 <- resultsAdultCA %>%
  as.tibble()  %>%
  rownames_to_column(var='target_id') %>%
  mutate(sig=ifelse(target_id %in% signifAdultCA[['target_id']], "FDR<0.05", "Not Sig")) %>%
  left_join(ttg, by = 'target_id')
r1 %>% ggplot(aes(log2FoldChange, -log10(padj))) + 
  geom_point(aes(col=sig)) + 
  scale_color_manual(values=c("red", "black")) +
#  coord_cartesian(ylim=c(0,10)) + 
  geom_text_repel(data=filter(r1, sig=='FDR<0.05'), aes(label=symbol))

Gene lists

Control - Fetal

ChildControl - Fetal

This table shows the 50 most significant genes (out of 10248) differentially expressed between control child samples and fetal samples.

genes_control_fetal_child <- signifChild %>%
  mutate(gene = text_spec(symbol, link = paste0('http://www.genecards.org/cgi-bin/carddisp.pl?gene=', symbol)),
         transcript = target_id) %>%
  dplyr::select(c('gene', 'transcript', 'log2FoldChange', 'padj', 'description', 'status', 'molecule_type')) %>%
  head(n = 50)
genes_control_fetal_child %>%
  kable() %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"), font_size = 6)

gene	transcript	log2FoldChange	padj	description	status	molecule_type
SOX11	NM_003108.3	-7.349298	0	SRY-box 11 (SOX11)	REVIEWED	mRNA
GPC2	NM_152742.2	-5.831508	0	glypican 2 (GPC2)	VALIDATED	mRNA
MEX3A	NM_001093725.1	-6.223953	0	mex-3 RNA binding family member A (MEX3A)	VALIDATED	mRNA
LMNB1	NM_005573.3	-5.958816	0	lamin B1 (LMNB1), transcript variant 1	REVIEWED	mRNA
KIF11	NM_004523.3	-5.032577	0	kinesin family member 11 (KIF11)	REVIEWED	mRNA
SOX4	NM_003107.2	-5.728428	0	SRY-box 4 (SOX4)	REVIEWED	mRNA
TMSB15A	NM_021992.2	-7.651072	0	thymosin beta 15a (TMSB15A)	VALIDATED	mRNA
GPR37L1	XM_011510158.2	6.780194	0	PREDICTED: Homo sapiens G protein-coupled receptor 37 like 1 (GPR37L1), transcript variant X1	MODEL	mRNA
LMNB1	NM_001198557.1	-6.272657	0	lamin B1 (LMNB1), transcript variant 2	REVIEWED	mRNA
IGIP	NM_001007189.1	3.815222	0	IgA inducing protein (IGIP)	VALIDATED	mRNA
MIR4461	NR_039666.1	14.481799	0	microRNA 4461 (MIR4461)	PROVISIONAL	microRNA
OMG	NM_002544.4	7.157673	0	oligodendrocyte myelin glycoprotein (OMG)	VALIDATED	mRNA
DCHS1	NM_003737.3	-3.096174	0	dachsous cadherin-related 1 (DCHS1)	REVIEWED	mRNA
SMC4	XM_011512312.2	-4.617128	0	PREDICTED: Homo sapiens structural maintenance of chromosomes 4 (SMC4), transcript variant X2	MODEL	mRNA
KNL1	XM_017022432.1	-5.821282	0	PREDICTED: Homo sapiens cancer susceptibility candidate 5 (CASC5), transcript variant X1	MODEL	mRNA
SPC25	NM_020675.3	-4.735962	0	SPC25, NDC80 kinetochore complex component (SPC25)	REVIEWED	mRNA
CDK1	NM_001786.4	-7.459224	0	cyclin dependent kinase 1 (CDK1), transcript variant 1	REVIEWED	mRNA
BCL2L2	NM_001199839.1	2.422108	0	BCL2 like 2 (BCL2L2), transcript variant 2	REVIEWED	mRNA
FGF1	NM_001257207.1	12.100679	0	fibroblast growth factor 1 (FGF1), transcript variant 9	REVIEWED	mRNA
PTGDS	NM_000954.5	12.971226	0	prostaglandin D2 synthase (PTGDS)	REVIEWED	mRNA
ASPM	NM_018136.4	-8.973773	0	abnormal spindle microtubule assembly (ASPM), transcript variant 1	REVIEWED	mRNA
ZBTB47	NM_145166.3	3.472201	0	zinc finger and BTB domain containing 47 (ZBTB47)	VALIDATED	mRNA
TPPP	XM_005248237.3	11.459962	0	PREDICTED: Homo sapiens tubulin polymerization promoting protein (TPPP), transcript variant X3	MODEL	mRNA
ZBED4	NM_014838.2	-2.546023	0	zinc finger BED-type containing 4 (ZBED4)	VALIDATED	mRNA
PLEKHB1	NM_001130035.1	6.346620	0	pleckstrin homology domain containing B1 (PLEKHB1), transcript variant 4	VALIDATED	mRNA
GPR37L1	NM_004767.3	7.914347	0	G protein-coupled receptor 37 like 1 (GPR37L1)	VALIDATED	mRNA
LGI3	NM_139278.2	11.287657	0	leucine rich repeat LGI family member 3 (LGI3)	VALIDATED	mRNA
TPPP	NM_007030.2	6.520413	0	tubulin polymerization promoting protein (TPPP)	VALIDATED	mRNA
CCNB2	NM_004701.3	-7.578472	0	cyclin B2 (CCNB2)	REVIEWED	mRNA
UBL3	NM_007106.3	2.274132	0	ubiquitin like 3 (UBL3)	VALIDATED	mRNA
ERMN	NM_020711.2	13.664215	0	ermin (ERMN), transcript variant 2	VALIDATED	mRNA
NDC80	NM_006101.2	-7.807808	0	NDC80, kinetochore complex component (NDC80)	VALIDATED	mRNA
S100A1	NM_006271.1	8.145612	0	S100 calcium binding protein A1 (S100A1)	REVIEWED	mRNA
HMGB3	NM_005342.3	-3.375052	0	high mobility group box 3 (HMGB3), transcript variant 2	REVIEWED	mRNA
MBP	XR_001753201.1	12.638850	0	PREDICTED: Homo sapiens myelin basic protein (MBP), transcript variant X2	MODEL	misc_RNA
HNRNPA0	NM_006805.3	-2.152540	0	heterogeneous nuclear ribonucleoprotein A0 (HNRNPA0)	REVIEWED	mRNA
LMNB2	NM_032737.3	-2.651268	0	lamin B2 (LMNB2)	REVIEWED	mRNA
NUSAP1	XM_005254430.4	-9.572857	0	PREDICTED: Homo sapiens nucleolar and spindle associated protein 1 (NUSAP1), transcript variant X5	MODEL	mRNA
RRM2	NM_001034.3	-6.321821	0	ribonucleotide reductase regulatory subunit M2 (RRM2), transcript variant 2	REVIEWED	mRNA
ZNF365	NM_014951.2	6.429411	0	zinc finger protein 365 (ZNF365), transcript variant A	REVIEWED	mRNA
SPOCK3	XM_017008257.1	10.848761	0	PREDICTED: Homo sapiens sparc/osteonectin, cwcv and kazal-like domains proteoglycan (testican) 3 (SPOCK3), transcript variant X1	MODEL	mRNA
PRNP	NM_001080123.2	4.128392	0	prion protein (PRNP), transcript variant 5	REVIEWED	mRNA
NIPAL3	NM_020448.4	11.523699	0	NIPA like domain containing 3 (NIPAL3), transcript variant 1	VALIDATED	mRNA
KIAA0930	NM_015264.1	10.790854	0	KIAA0930 (KIAA0930), transcript variant 1	VALIDATED	mRNA
ESCO2	NM_001017420.2	-6.652677	0	establishment of sister chromatid cohesion N-acetyltransferase 2 (ESCO2)	REVIEWED	mRNA
ENDOD1	NM_015036.2	5.798479	0	endonuclease domain containing 1 (ENDOD1)	VALIDATED	mRNA
TACC3	NM_006342.2	-4.637760	0	transforming acidic coiled-coil containing protein 3 (TACC3)	REVIEWED	mRNA
H1F0	NM_005318.3	-2.744947	0	H1 histone family member 0 (H1F0)	REVIEWED	mRNA
ALDH1A1	NM_000689.4	10.679490	0	aldehyde dehydrogenase 1 family member A1 (ALDH1A1)	REVIEWED	mRNA
LDLRAD4	NM_001003674.3	8.798093	0	low density lipoprotein receptor class A domain containing 4 (LDLRAD4), transcript variant c1	VALIDATED	mRNA

AdultControl - Fetal

This table shows the 50 most significant genes (out of 11265) differentially expressed between control adult samples and fetal samples.

genes_control_fetal_adult <- signifAdult %>%
  mutate(gene = text_spec(symbol, link = paste0('http://www.genecards.org/cgi-bin/carddisp.pl?gene=', symbol)),
         transcript = target_id) %>%
  dplyr::select(c('gene', 'transcript', 'log2FoldChange', 'padj', 'description', 'status', 'molecule_type')) %>%
  head(n = 50)
genes_control_fetal_adult %>%
  kable() %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"), font_size = 6)

gene	transcript	log2FoldChange	padj	description	status	molecule_type
SOX11	NM_003108.3	-8.097169	0	SRY-box 11 (SOX11)	REVIEWED	mRNA
MEX3A	NM_001093725.1	-6.578792	0	mex-3 RNA binding family member A (MEX3A)	VALIDATED	mRNA
SOX4	NM_003107.2	-6.728145	0	SRY-box 4 (SOX4)	REVIEWED	mRNA
GPC2	NM_152742.2	-5.070847	0	glypican 2 (GPC2)	VALIDATED	mRNA
LMNB1	NM_005573.3	-6.490096	0	lamin B1 (LMNB1), transcript variant 1	REVIEWED	mRNA
KIF11	NM_004523.3	-4.945373	0	kinesin family member 11 (KIF11)	REVIEWED	mRNA
DCHS1	NM_003737.3	-3.864829	0	dachsous cadherin-related 1 (DCHS1)	REVIEWED	mRNA
GPR37L1	XM_011510158.2	6.460557	0	PREDICTED: Homo sapiens G protein-coupled receptor 37 like 1 (GPR37L1), transcript variant X1	MODEL	mRNA
LMNB1	NM_001198557.1	-6.686956	0	lamin B1 (LMNB1), transcript variant 2	REVIEWED	mRNA
OMG	NM_002544.4	7.999721	0	oligodendrocyte myelin glycoprotein (OMG)	VALIDATED	mRNA
IGIP	NM_001007189.1	4.091410	0	IgA inducing protein (IGIP)	VALIDATED	mRNA
TMSB15A	NM_021992.2	-8.533119	0	thymosin beta 15a (TMSB15A)	VALIDATED	mRNA
SMC4	XM_011512312.2	-5.011910	0	PREDICTED: Homo sapiens structural maintenance of chromosomes 4 (SMC4), transcript variant X2	MODEL	mRNA
HMGB3	NM_005342.3	-4.076148	0	high mobility group box 3 (HMGB3), transcript variant 2	REVIEWED	mRNA
HNRNPA0	NM_006805.3	-2.569509	0	heterogeneous nuclear ribonucleoprotein A0 (HNRNPA0)	REVIEWED	mRNA
UBL3	NM_007106.3	2.625024	0	ubiquitin like 3 (UBL3)	VALIDATED	mRNA
ENDOD1	NM_015036.2	7.089989	0	endonuclease domain containing 1 (ENDOD1)	VALIDATED	mRNA
BCL2L2	NM_001199839.1	2.552130	0	BCL2 like 2 (BCL2L2), transcript variant 2	REVIEWED	mRNA
S100A1	NM_006271.1	9.358662	0	S100 calcium binding protein A1 (S100A1)	REVIEWED	mRNA
FGF1	NM_001257207.1	12.592264	0	fibroblast growth factor 1 (FGF1), transcript variant 9	REVIEWED	mRNA
PLEKHB1	NM_001130035.1	6.910400	0	pleckstrin homology domain containing B1 (PLEKHB1), transcript variant 4	VALIDATED	mRNA
TPPP	NM_007030.2	7.137853	0	tubulin polymerization promoting protein (TPPP)	VALIDATED	mRNA
PTGDS	NM_000954.5	13.303835	0	prostaglandin D2 synthase (PTGDS)	REVIEWED	mRNA
SPOCK3	XM_017008257.1	12.397706	0	PREDICTED: Homo sapiens sparc/osteonectin, cwcv and kazal-like domains proteoglycan (testican) 3 (SPOCK3), transcript variant X1	MODEL	mRNA
DBN1	NM_004395.3	-4.630562	0	drebrin 1 (DBN1), transcript variant 1	REVIEWED	mRNA
ERMN	NM_020711.2	14.938566	0	ermin (ERMN), transcript variant 2	VALIDATED	mRNA
MBP	XR_001753201.1	13.972643	0	PREDICTED: Homo sapiens myelin basic protein (MBP), transcript variant X2	MODEL	misc_RNA
LGI3	NM_139278.2	12.050597	0	leucine rich repeat LGI family member 3 (LGI3)	VALIDATED	mRNA
LMNB2	NM_032737.3	-2.961563	0	lamin B2 (LMNB2)	REVIEWED	mRNA
KNL1	XM_017022432.1	-5.541103	0	PREDICTED: Homo sapiens cancer susceptibility candidate 5 (CASC5), transcript variant X1	MODEL	mRNA
ASPM	NM_018136.4	-8.236651	0	abnormal spindle microtubule assembly (ASPM), transcript variant 1	REVIEWED	mRNA
MIR4461	NR_039666.1	13.373719	0	microRNA 4461 (MIR4461)	PROVISIONAL	microRNA
TF	NM_001063.3	11.151690	0	transferrin (TF)	REVIEWED	mRNA
CDK1	NM_001786.4	-7.247968	0	cyclin dependent kinase 1 (CDK1), transcript variant 1	REVIEWED	mRNA
GPR37L1	NM_004767.3	8.065756	0	G protein-coupled receptor 37 like 1 (GPR37L1)	VALIDATED	mRNA
LDLRAD4	NM_001003674.3	9.785637	0	low density lipoprotein receptor class A domain containing 4 (LDLRAD4), transcript variant c1	VALIDATED	mRNA
ZBTB47	NM_145166.3	3.373056	0	zinc finger and BTB domain containing 47 (ZBTB47)	VALIDATED	mRNA
TPPP	XM_005248237.3	11.351597	0	PREDICTED: Homo sapiens tubulin polymerization promoting protein (TPPP), transcript variant X3	MODEL	mRNA
MOBP	NR_003090.2	11.274860	0	myelin-associated oligodendrocyte basic protein (MOBP), transcript variant 4	VALIDATED	non-coding RNA
KIAA0930	NM_015264.1	11.619154	0	KIAA0930 (KIAA0930), transcript variant 1	VALIDATED	mRNA
ENPP4	NM_014936.4	5.494009	0	ectonucleotide pyrophosphatase/phosphodiesterase 4 (putative) (ENPP4)	VALIDATED	mRNA
TACC3	NM_006342.2	-5.529844	0	transforming acidic coiled-coil containing protein 3 (TACC3)	REVIEWED	mRNA
RRM2	NM_001034.3	-7.069335	0	ribonucleotide reductase regulatory subunit M2 (RRM2), transcript variant 2	REVIEWED	mRNA
ZNF365	NM_014951.2	6.712768	0	zinc finger protein 365 (ZNF365), transcript variant A	REVIEWED	mRNA
NONO	NM_007363.4	-2.181021	0	non-POU domain containing octamer binding (NONO), transcript variant 2	REVIEWED	mRNA
H1F0	NM_005318.3	-2.948282	0	H1 histone family member 0 (H1F0)	REVIEWED	mRNA
NIPAL3	NM_020448.4	12.074509	0	NIPA like domain containing 3 (NIPAL3), transcript variant 1	VALIDATED	mRNA
CNP	NM_033133.4	6.946693	0	2’,3’-cyclic nucleotide 3’ phosphodiesterase (CNP), transcript variant 1	VALIDATED	mRNA
DESI1	NM_015704.2	1.891739	0	desumoylating isopeptidase 1 (DESI1)	VALIDATED	mRNA
LOC105379481	XR_951069.2	12.748886	0	PREDICTED: Homo sapiens uncharacterized LOC105379481 (LOC105379481)	MODEL	ncRNA

Autism - Fetal

ChildAutism - Fetal

This table shows the 50 most significant genes (out of 11626) differentially expressed between autism child samples and fetal samples.

genes_autism_fetal_child <- signifChildA %>%
  mutate(gene = text_spec(symbol, link = paste0('http://www.genecards.org/cgi-bin/carddisp.pl?gene=', symbol)),
         transcript = target_id) %>%
  dplyr::select(c('gene', 'transcript', 'log2FoldChange', 'padj', 'description', 'status', 'molecule_type')) %>%
  head(n = 50)
genes_autism_fetal_child %>%
  kable() %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"), font_size = 6)

gene	transcript	log2FoldChange	padj	description	status	molecule_type
SOX11	NM_003108.3	-6.801260	0	SRY-box 11 (SOX11)	REVIEWED	mRNA
GPC2	NM_152742.2	-5.918823	0	glypican 2 (GPC2)	VALIDATED	mRNA
MEX3A	NM_001093725.1	-6.194712	0	mex-3 RNA binding family member A (MEX3A)	VALIDATED	mRNA
LMNB1	NM_005573.3	-6.235398	0	lamin B1 (LMNB1), transcript variant 1	REVIEWED	mRNA
KIF11	NM_004523.3	-5.564826	0	kinesin family member 11 (KIF11)	REVIEWED	mRNA
SOX4	NM_003107.2	-5.559113	0	SRY-box 4 (SOX4)	REVIEWED	mRNA
GPR37L1	XM_011510158.2	6.953017	0	PREDICTED: Homo sapiens G protein-coupled receptor 37 like 1 (GPR37L1), transcript variant X1	MODEL	mRNA
TMSB15A	NM_021992.2	-7.984293	0	thymosin beta 15a (TMSB15A)	VALIDATED	mRNA
DCHS1	NM_003737.3	-3.592684	0	dachsous cadherin-related 1 (DCHS1)	REVIEWED	mRNA
SMC4	XM_011512312.2	-5.333317	0	PREDICTED: Homo sapiens structural maintenance of chromosomes 4 (SMC4), transcript variant X2	MODEL	mRNA
LMNB1	NM_001198557.1	-6.112901	0	lamin B1 (LMNB1), transcript variant 2	REVIEWED	mRNA
IGIP	NM_001007189.1	3.858878	0	IgA inducing protein (IGIP)	VALIDATED	mRNA
OMG	NM_002544.4	7.287363	0	oligodendrocyte myelin glycoprotein (OMG)	VALIDATED	mRNA
MIR4461	NR_039666.1	14.513115	0	microRNA 4461 (MIR4461)	PROVISIONAL	microRNA
KNL1	XM_017022432.1	-6.290562	0	PREDICTED: Homo sapiens cancer susceptibility candidate 5 (CASC5), transcript variant X1	MODEL	mRNA
BCL2L2	NM_001199839.1	2.455968	0	BCL2 like 2 (BCL2L2), transcript variant 2	REVIEWED	mRNA
NUSAP1	XM_005254430.4	-7.859581	0	PREDICTED: Homo sapiens nucleolar and spindle associated protein 1 (NUSAP1), transcript variant X5	MODEL	mRNA
SPC25	NM_020675.3	-5.147332	0	SPC25, NDC80 kinetochore complex component (SPC25)	REVIEWED	mRNA
QSER1	NM_001076786.2	-3.338748	0	glutamine and serine rich 1 (QSER1)	VALIDATED	mRNA
ZBED4	NM_014838.2	-2.644819	0	zinc finger BED-type containing 4 (ZBED4)	VALIDATED	mRNA
FOXO3B	NR_026718.1	-3.683149	0	forkhead box O3B pseudogene (FOXO3B)	PROVISIONAL	non-coding RNA
CDK1	NM_001786.4	-7.560371	0	cyclin dependent kinase 1 (CDK1), transcript variant 1	REVIEWED	mRNA
ZBTB47	NM_145166.3	3.445832	0	zinc finger and BTB domain containing 47 (ZBTB47)	VALIDATED	mRNA
ELAVL2	NM_001351477.1	-8.889964	0	ELAV like RNA binding protein 2 (ELAVL2), transcript variant 26	REVIEWED	mRNA
S100A1	NM_006271.1	8.807386	0	S100 calcium binding protein A1 (S100A1)	REVIEWED	mRNA
PLEKHB1	NM_001130035.1	6.239793	0	pleckstrin homology domain containing B1 (PLEKHB1), transcript variant 4	VALIDATED	mRNA
HNRNPA0	NM_006805.3	-2.194815	0	heterogeneous nuclear ribonucleoprotein A0 (HNRNPA0)	REVIEWED	mRNA
ASPM	NM_018136.4	-9.287765	0	abnormal spindle microtubule assembly (ASPM), transcript variant 1	REVIEWED	mRNA
ATP1B1	NM_001677.3	4.446844	0	ATPase Na+/K+ transporting subunit beta 1 (ATP1B1)	REVIEWED	mRNA
PTGDS	NM_000954.5	12.567492	0	prostaglandin D2 synthase (PTGDS)	REVIEWED	mRNA
RRM2	NM_001034.3	-6.725265	0	ribonucleotide reductase regulatory subunit M2 (RRM2), transcript variant 2	REVIEWED	mRNA
GPR37L1	NM_004767.3	8.043598	0	G protein-coupled receptor 37 like 1 (GPR37L1)	VALIDATED	mRNA
ZNF365	NM_014951.2	6.661035	0	zinc finger protein 365 (ZNF365), transcript variant A	REVIEWED	mRNA
HNRNPA1	NM_002136.3	-2.532387	0	heterogeneous nuclear ribonucleoprotein A1 (HNRNPA1), transcript variant 1	REVIEWED	mRNA
CCNB2	NM_004701.3	-6.754708	0	cyclin B2 (CCNB2)	REVIEWED	mRNA
TPPP	NM_007030.2	6.225312	0	tubulin polymerization promoting protein (TPPP)	VALIDATED	mRNA
PINK1	NM_032409.2	3.107355	0	PTEN induced putative kinase 1 (PINK1)	REVIEWED	mRNA
EOMES	NM_001278183.1	-10.635396	0	eomesodermin (EOMES), transcript variant 3	REVIEWED	mRNA
FGF1	NM_001257207.1	11.471307	0	fibroblast growth factor 1 (FGF1), transcript variant 9	REVIEWED	mRNA
PEBP1	NM_002567.3	2.431571	0	phosphatidylethanolamine binding protein 1 (PEBP1)	REVIEWED	mRNA
LGI3	NM_139278.2	11.319490	0	leucine rich repeat LGI family member 3 (LGI3)	VALIDATED	mRNA
TACC3	NM_006342.2	-4.877290	0	transforming acidic coiled-coil containing protein 3 (TACC3)	REVIEWED	mRNA
PRNP	NM_001080123.2	4.106136	0	prion protein (PRNP), transcript variant 5	REVIEWED	mRNA
H1F0	NM_005318.3	-2.783194	0	H1 histone family member 0 (H1F0)	REVIEWED	mRNA
NDC80	NM_006101.2	-7.569346	0	NDC80, kinetochore complex component (NDC80)	VALIDATED	mRNA
LMNB2	NM_032737.3	-2.533771	0	lamin B2 (LMNB2)	REVIEWED	mRNA
SPOCK3	XM_017008257.1	11.274727	0	PREDICTED: Homo sapiens sparc/osteonectin, cwcv and kazal-like domains proteoglycan (testican) 3 (SPOCK3), transcript variant X1	MODEL	mRNA
ERMN	NM_020711.2	13.266357	0	ermin (ERMN), transcript variant 2	VALIDATED	mRNA
TF	NM_001063.3	9.623189	0	transferrin (TF)	REVIEWED	mRNA
HCN2	NM_001194.3	7.676608	0	hyperpolarization activated cyclic nucleotide gated potassium and sodium channel 2 (HCN2)	REVIEWED	mRNA

AdultAutism - Fetal

This table shows the 50 most significant genes (out of 11089) differentially expressed between autism adult samples and fetal samples.

genes_autism_fetal_adult <- signifAdultA %>%
  mutate(gene = text_spec(symbol, link = paste0('http://www.genecards.org/cgi-bin/carddisp.pl?gene=', symbol)),
         transcript = target_id) %>%
  dplyr::select(c('gene', 'transcript', 'log2FoldChange', 'padj', 'description', 'status', 'molecule_type')) %>%
  head(n = 50)
genes_autism_fetal_adult %>%
  kable() %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"), font_size = 6)

gene	transcript	log2FoldChange	padj	description	status	molecule_type
SOX11	NM_003108.3	-7.268496	0	SRY-box 11 (SOX11)	REVIEWED	mRNA
MEX3A	NM_001093725.1	-6.540832	0	mex-3 RNA binding family member A (MEX3A)	VALIDATED	mRNA
SOX4	NM_003107.2	-6.706596	0	SRY-box 4 (SOX4)	REVIEWED	mRNA
GPC2	NM_152742.2	-5.405610	0	glypican 2 (GPC2)	VALIDATED	mRNA
LMNB1	NM_005573.3	-6.387707	0	lamin B1 (LMNB1), transcript variant 1	REVIEWED	mRNA
KIF11	NM_004523.3	-5.135591	0	kinesin family member 11 (KIF11)	REVIEWED	mRNA
SNORA109	NR_132964.1	14.999271	0	small nucleolar RNA, H/ACA box 109 (SNORA109)	VALIDATED	small nucleolar RNA
GPR37L1	XM_011510158.2	6.418046	0	PREDICTED: Homo sapiens G protein-coupled receptor 37 like 1 (GPR37L1), transcript variant X1	MODEL	mRNA
LMNB1	NM_001198557.1	-6.591551	0	lamin B1 (LMNB1), transcript variant 2	REVIEWED	mRNA
IGIP	NM_001007189.1	4.085882	0	IgA inducing protein (IGIP)	VALIDATED	mRNA
SMC4	XM_011512312.2	-5.262400	0	PREDICTED: Homo sapiens structural maintenance of chromosomes 4 (SMC4), transcript variant X2	MODEL	mRNA
DCHS1	NM_003737.3	-3.399609	0	dachsous cadherin-related 1 (DCHS1)	REVIEWED	mRNA
HNRNPA0	NM_006805.3	-2.701007	0	heterogeneous nuclear ribonucleoprotein A0 (HNRNPA0)	REVIEWED	mRNA
OMG	NM_002544.4	7.561685	0	oligodendrocyte myelin glycoprotein (OMG)	VALIDATED	mRNA
HMGB3	NM_005342.3	-4.084812	0	high mobility group box 3 (HMGB3), transcript variant 2	REVIEWED	mRNA
S100A1	NM_006271.1	9.010194	0	S100 calcium binding protein A1 (S100A1)	REVIEWED	mRNA
TPPP	NM_007030.2	7.041179	0	tubulin polymerization promoting protein (TPPP)	VALIDATED	mRNA
PTGDS	NM_000954.5	13.046271	0	prostaglandin D2 synthase (PTGDS)	REVIEWED	mRNA
TMSB15A	NM_021992.2	-9.533347	0	thymosin beta 15a (TMSB15A)	VALIDATED	mRNA
PLEKHB1	NM_001130035.1	6.712008	0	pleckstrin homology domain containing B1 (PLEKHB1), transcript variant 4	VALIDATED	mRNA
BCL2L2	NM_001199839.1	2.409943	0	BCL2 like 2 (BCL2L2), transcript variant 2	REVIEWED	mRNA
MIR4461	NR_039666.1	13.390095	0	microRNA 4461 (MIR4461)	PROVISIONAL	microRNA
ASPM	NM_018136.4	-8.307580	0	abnormal spindle microtubule assembly (ASPM), transcript variant 1	REVIEWED	mRNA
SPOCK3	XM_017008257.1	12.103434	0	PREDICTED: Homo sapiens sparc/osteonectin, cwcv and kazal-like domains proteoglycan (testican) 3 (SPOCK3), transcript variant X1	MODEL	mRNA
KNL1	XM_017022432.1	-5.445786	0	PREDICTED: Homo sapiens cancer susceptibility candidate 5 (CASC5), transcript variant X1	MODEL	mRNA
CDK1	NM_001786.4	-7.661187	0	cyclin dependent kinase 1 (CDK1), transcript variant 1	REVIEWED	mRNA
FGF1	NM_001257207.1	11.871241	0	fibroblast growth factor 1 (FGF1), transcript variant 9	REVIEWED	mRNA
LGI3	NM_139278.2	11.801906	0	leucine rich repeat LGI family member 3 (LGI3)	VALIDATED	mRNA
KIAA0930	NM_015264.1	12.021455	0	KIAA0930 (KIAA0930), transcript variant 1	VALIDATED	mRNA
H1F0	NM_005318.3	-3.108088	0	H1 histone family member 0 (H1F0)	REVIEWED	mRNA
MIR6723	NR_106781.1	14.605592	0	microRNA 6723 (MIR6723)	PROVISIONAL	microRNA
LMNB2	NM_032737.3	-2.832149	0	lamin B2 (LMNB2)	REVIEWED	mRNA
ERMN	NM_020711.2	14.190130	0	ermin (ERMN), transcript variant 2	VALIDATED	mRNA
MBP	XR_001753201.1	13.264769	0	PREDICTED: Homo sapiens myelin basic protein (MBP), transcript variant X2	MODEL	misc_RNA
TPPP	XM_005248237.3	11.385896	0	PREDICTED: Homo sapiens tubulin polymerization promoting protein (TPPP), transcript variant X3	MODEL	mRNA
ENDOD1	NM_015036.2	6.364347	0	endonuclease domain containing 1 (ENDOD1)	VALIDATED	mRNA
ZBTB47	NM_145166.3	3.373055	0	zinc finger and BTB domain containing 47 (ZBTB47)	VALIDATED	mRNA
SPC25	NM_020675.3	-4.111927	0	SPC25, NDC80 kinetochore complex component (SPC25)	REVIEWED	mRNA
TF	NM_001063.3	10.484396	0	transferrin (TF)	REVIEWED	mRNA
HNRNPA1	NM_002136.3	-2.632144	0	heterogeneous nuclear ribonucleoprotein A1 (HNRNPA1), transcript variant 1	REVIEWED	mRNA
GPR37L1	NM_004767.3	7.832109	0	G protein-coupled receptor 37 like 1 (GPR37L1)	VALIDATED	mRNA
UBL3	NM_007106.3	2.254224	0	ubiquitin like 3 (UBL3)	VALIDATED	mRNA
LOC105379481	XR_951069.2	12.675918	0	PREDICTED: Homo sapiens uncharacterized LOC105379481 (LOC105379481)	MODEL	ncRNA
NONO	NM_007363.4	-2.124006	0	non-POU domain containing octamer binding (NONO), transcript variant 2	REVIEWED	mRNA
TACC3	NM_006342.2	-5.062962	0	transforming acidic coiled-coil containing protein 3 (TACC3)	REVIEWED	mRNA
MOBP	NR_003090.2	10.636680	0	myelin-associated oligodendrocyte basic protein (MOBP), transcript variant 4	VALIDATED	non-coding RNA
PRNP	NM_001080123.2	4.186647	0	prion protein (PRNP), transcript variant 5	REVIEWED	mRNA
LDLRAD4	NM_001003674.3	9.111582	0	low density lipoprotein receptor class A domain containing 4 (LDLRAD4), transcript variant c1	VALIDATED	mRNA
PTTG1	NM_004219.3	-6.222015	0	pituitary tumor-transforming 1 (PTTG1), transcript variant 2	REVIEWED	mRNA
ZNF365	NM_014951.2	6.381353	0	zinc finger protein 365 (ZNF365), transcript variant A	REVIEWED	mRNA

Control - Autism

Child

This table shows the 32 most significant genes differentially expressed between control child samples and autism child samples.

genes_control_autism_child <- signifChildCA %>%
  mutate(gene = text_spec(symbol, link = paste0('http://www.genecards.org/cgi-bin/carddisp.pl?gene=', symbol)),
         transcript = target_id) %>%
  dplyr::select(c('gene', 'transcript', 'log2FoldChange', 'padj', 'description', 'status', 'molecule_type')) %>%
  head(n = 50)
genes_control_autism_child %>%
  kable() %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"), font_size = 6)

gene	transcript	log2FoldChange	padj	description	status	molecule_type
H2AFY	NM_001040158.1	-8.025760	0.0028433	H2A histone family member Y (H2AFY), transcript variant 4	REVIEWED	mRNA
ZSCAN25	NM_001350984.1	5.127925	0.0030546	zinc finger and SCAN domain containing 25 (ZSCAN25), transcript variant 7	VALIDATED	mRNA
PIK3R3	NM_003629.3	7.601205	0.0030546	phosphoinositide-3-kinase regulatory subunit 3 (PIK3R3), transcript variant 1	REVIEWED	mRNA
ZNF195	XM_011520352.2	6.694985	0.0030546	PREDICTED: Homo sapiens zinc finger protein 195 (ZNF195), transcript variant X7	MODEL	mRNA
SUSD5	XM_017006137.1	6.711799	0.0035603	PREDICTED: Homo sapiens sushi domain containing 5 (SUSD5), transcript variant X2	MODEL	mRNA
ACVR1C	NM_145259.2	8.014360	0.0039174	activin A receptor type 1C (ACVR1C), transcript variant 1	VALIDATED	mRNA
BUB1B	NM_001211.5	3.315151	0.0063885	BUB1 mitotic checkpoint serine/threonine kinase B (BUB1B)	REVIEWED	mRNA
PCM1	XM_017013484.1	7.884200	0.0063885	PREDICTED: Homo sapiens pericentriolar material 1 (PCM1), transcript variant X23	MODEL	mRNA
ELAVL2	XM_011517776.2	5.712591	0.0069766	PREDICTED: Homo sapiens ELAV like neuron-specific RNA binding protein 2 (ELAVL2), transcript variant X1	MODEL	mRNA
ZFP64	XM_005260449.1	-7.105792	0.0084906	PREDICTED: Homo sapiens ZFP64 zinc finger protein (ZFP64), transcript variant X6	MODEL	mRNA
MYOT	XM_017010061.1	-5.617699	0.0137692	PREDICTED: Homo sapiens myotilin (MYOT), transcript variant X2	MODEL	mRNA
RGL1	XM_017000756.1	6.326049	0.0155666	PREDICTED: Homo sapiens ral guanine nucleotide dissociation stimulator like 1 (RGL1), transcript variant X3	MODEL	mRNA
TCF4	XM_017025937.1	7.196867	0.0155666	PREDICTED: Homo sapiens transcription factor 4 (TCF4), transcript variant X5	MODEL	mRNA
ZNF160	NM_001322136.1	-6.555408	0.0156208	zinc finger protein 160 (ZNF160), transcript variant 12	REVIEWED	mRNA
CHD3	XM_017024064.1	7.552134	0.0156208	PREDICTED: Homo sapiens chromodomain helicase DNA binding protein 3 (CHD3), transcript variant X9	MODEL	mRNA
EXOC1	NM_018261.3	-6.547517	0.0162899	exocyst complex component 1 (EXOC1), transcript variant 1	REVIEWED	mRNA
TTR	NM_000371.3	7.428806	0.0207366	transthyretin (TTR)	REVIEWED	mRNA
PELI3	NM_001243136.1	-8.116381	0.0207366	pellino E3 ubiquitin protein ligase family member 3 (PELI3), transcript variant 4	REVIEWED	mRNA
PAG1	NM_018440.3	7.766469	0.0207366	phosphoprotein membrane anchor with glycosphingolipid microdomains 1 (PAG1)	REVIEWED	mRNA
LRRC75A-AS1	NR_027159.1	-5.821114	0.0207366	LRRC75A antisense RNA 1 (LRRC75A-AS1), transcript variant 3	PREDICTED	long non-coding RNA
PLPP5	XM_017013905.1	5.051513	0.0207366	PREDICTED: Homo sapiens phospholipid phosphatase 5 (PLPP5), transcript variant X10	MODEL	mRNA
YPEL2	XM_017024621.1	6.488686	0.0207366	PREDICTED: Homo sapiens yippee like 2 (YPEL2), transcript variant X1	MODEL	mRNA
OSBPL2	XM_017028170.1	6.644400	0.0207366	PREDICTED: Homo sapiens oxysterol binding protein like 2 (OSBPL2), transcript variant X8	MODEL	mRNA
NEUROD4	NM_021191.2	5.695382	0.0261895	neuronal differentiation 4 (NEUROD4)	VALIDATED	mRNA
STK38L	XM_017019036.1	7.354445	0.0334952	PREDICTED: Homo sapiens serine/threonine kinase 38 like (STK38L), transcript variant X2	MODEL	mRNA
MCM7	NM_001278595.1	-6.148432	0.0341248	minichromosome maintenance complex component 7 (MCM7), transcript variant 3	REVIEWED	mRNA
AP2S1	NM_001301081.1	-6.879410	0.0369278	adaptor related protein complex 2 sigma 1 subunit (AP2S1), transcript variant 5	REVIEWED	mRNA
RAB11FIP1	NM_025151.4	2.165864	0.0369278	RAB11 family interacting protein 1 (RAB11FIP1), transcript variant 1	VALIDATED	mRNA
SFRP1	NM_003012.4	2.112847	0.0385669	secreted frizzled related protein 1 (SFRP1)	REVIEWED	mRNA
A4GALT	XM_005261647.2	-6.759707	0.0428226	PREDICTED: Homo sapiens alpha 1,4-galactosyltransferase (A4GALT), transcript variant X4	MODEL	mRNA
TAGAP	NM_152133.2	5.808768	0.0497265	T-cell activation RhoGTPase activating protein (TAGAP), transcript variant 1	REVIEWED	mRNA
LMBR1	XR_001744847.1	-4.560443	0.0497265	PREDICTED: Homo sapiens limb development membrane protein 1 (LMBR1), transcript variant X1	MODEL	misc_RNA

Adult

This table shows the 50 most significant genes differentially expressed between control adult samples and autism adult samples.

genes_control_autism_adult <- signifAdultCA %>%
  mutate(gene = text_spec(symbol, link = paste0('http://www.genecards.org/cgi-bin/carddisp.pl?gene=', symbol)),
         transcript = target_id) %>%
  dplyr::select(c('gene', 'transcript', 'log2FoldChange', 'padj', 'description', 'status', 'molecule_type')) %>%
  head(n = 50)
genes_control_autism_adult %>%
  kable() %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"), font_size = 6)

gene	transcript	log2FoldChange	padj	description	status	molecule_type
FHOD3	XM_005258355.1	-8.217616	0.0000000	PREDICTED: Homo sapiens formin homology 2 domain containing 3 (FHOD3), transcript variant X13	MODEL	mRNA
LOC107986119	XR_001750192.1	7.763706	0.0000087	PREDICTED: Homo sapiens uncharacterized LOC107986119 (LOC107986119), transcript variant X1	MODEL	ncRNA
DLGAP1	NM_001242764.1	-7.603438	0.0000505	DLG associated protein 1 (DLGAP1), transcript variant 6	VALIDATED	mRNA
AFF2	NM_001170628.1	-5.376843	0.0011518	AF4/FMR2 family member 2 (AFF2), transcript variant 6	REVIEWED	mRNA
CSE1L	NR_045796.1	-7.724335	0.0018763	chromosome segregation 1 like (CSE1L), transcript variant 3	REVIEWED	non-coding RNA
PRKCE	XM_006712050.3	-7.724396	0.0019385	PREDICTED: Homo sapiens protein kinase C epsilon (PRKCE), transcript variant X17	MODEL	mRNA
LOC107985374	XR_001735766.1	6.458495	0.0025999	PREDICTED: Homo sapiens uncharacterized LOC107985374 (LOC107985374)	MODEL	ncRNA
CHD3	XM_017024066.1	7.378109	0.0028590	PREDICTED: Homo sapiens chromodomain helicase DNA binding protein 3 (CHD3), transcript variant X11	MODEL	mRNA
LOC102724788	XM_006724935.3	-8.186769	0.0028638	PREDICTED: Homo sapiens proline dehydrogenase 1, mitochondrial (LOC102724788), transcript variant X1	MODEL	mRNA
ZNF821	XM_017023411.1	-6.516153	0.0028638	PREDICTED: Homo sapiens zinc finger protein 821 (ZNF821), transcript variant X4	MODEL	mRNA
OGT	XM_017029907.1	-7.653633	0.0034868	PREDICTED: Homo sapiens O-linked N-acetylglucosamine (GlcNAc) transferase (OGT), transcript variant X1	MODEL	mRNA
LOC107986119	XR_001740863.1	-7.333702	0.0036826	PREDICTED: Homo sapiens uncharacterized LOC107986119 (LOC107986119), transcript variant X1	MODEL	ncRNA
ANO1	NM_018043.5	5.930273	0.0047391	anoctamin 1 (ANO1), transcript variant 1	VALIDATED	mRNA
GRIA1	XM_017009393.1	-6.981561	0.0047391	PREDICTED: Homo sapiens glutamate ionotropic receptor AMPA type subunit 1 (GRIA1), transcript variant X3	MODEL	mRNA
EIF3C	XM_017023814.1	7.513897	0.0047391	PREDICTED: Homo sapiens eukaryotic translation initiation factor 3 subunit C (EIF3C), transcript variant X1	MODEL	mRNA
PASK	XM_011510835.1	-7.389713	0.0060558	PREDICTED: Homo sapiens PAS domain containing serine/threonine kinase (PASK), transcript variant X14	MODEL	mRNA
ATAD2	XM_011516996.2	-2.870322	0.0084250	PREDICTED: Homo sapiens ATPase family, AAA domain containing 2 (ATAD2), transcript variant X3	MODEL	mRNA
SCAF11	XM_017020221.1	6.422783	0.0100051	PREDICTED: Homo sapiens SR-related CTD associated factor 11 (SCAF11), transcript variant X8	MODEL	mRNA
WNK2	XM_011518936.2	-6.311372	0.0103810	PREDICTED: Homo sapiens WNK lysine deficient protein kinase 2 (WNK2), transcript variant X22	MODEL	mRNA
TMEM268	XM_017014429.1	5.870467	0.0103810	PREDICTED: Homo sapiens chromosome 9 open reading frame 91 (C9orf91), transcript variant X3	MODEL	mRNA
LSM6	XR_001741100.1	6.039595	0.0103810	PREDICTED: Homo sapiens LSM6 homolog, U6 small nuclear RNA and mRNA degradation associated (LSM6), transcript variant X3	MODEL	misc_RNA
NCAPG2	NM_017760.6	7.068658	0.0105529	non-SMC condensin II complex subunit G2 (NCAPG2), transcript variant 1	REVIEWED	mRNA
SETD2	NM_001349370.1	-7.886412	0.0116732	SET domain containing 2 (SETD2), transcript variant 2	REVIEWED	mRNA
LOC101929097	XM_005259695.2	-8.034844	0.0116732	PREDICTED: Homo sapiens uncharacterized LOC101929097 (LOC101929097)	MODEL	mRNA
CHD3	XM_017024064.1	7.747000	0.0117789	PREDICTED: Homo sapiens chromodomain helicase DNA binding protein 3 (CHD3), transcript variant X9	MODEL	mRNA
ACVR1C	NM_001111033.1	-7.267053	0.0134341	activin A receptor type 1C (ACVR1C), transcript variant 4	VALIDATED	mRNA
UBA1	XM_017029780.1	-5.980508	0.0134341	PREDICTED: Homo sapiens ubiquitin like modifier activating enzyme 1 (UBA1), transcript variant X6	MODEL	mRNA
CGN	XM_005245365.4	7.766203	0.0136224	PREDICTED: Homo sapiens cingulin (CGN), transcript variant X1	MODEL	mRNA
H2AFY	XM_011543734.2	6.261827	0.0136224	PREDICTED: Homo sapiens H2A histone family member Y (H2AFY), transcript variant X25	MODEL	mRNA
TAB2	NM_001292035.2	-4.660715	0.0146400	TGF-beta activated kinase 1/MAP3K7 binding protein 2 (TAB2), transcript variant 4	REVIEWED	mRNA
LIN52	NM_001024674.2	6.294265	0.0150929	lin-52 DREAM MuvB core complex component (LIN52)	VALIDATED	mRNA
PITPNM3	NM_001165966.1	-7.579690	0.0170745	PITPNM family member 3 (PITPNM3), transcript variant 2	REVIEWED	mRNA
CEP170	XM_011544343.2	6.667994	0.0252273	PREDICTED: Homo sapiens centrosomal protein 170 (CEP170), transcript variant X23	MODEL	mRNA
STARD8	XM_011531069.2	6.843532	0.0274375	PREDICTED: Homo sapiens StAR related lipid transfer domain containing 8 (STARD8), transcript variant X2	MODEL	mRNA
TNFRSF10D	NM_003840.4	-2.259592	0.0323857	TNF receptor superfamily member 10d (TNFRSF10D)	REVIEWED	mRNA
SPAST	NM_014946.3	-4.304759	0.0360086	spastin (SPAST), transcript variant 1	REVIEWED	mRNA
TPX2	XM_011528697.2	5.036863	0.0360086	PREDICTED: Homo sapiens TPX2, microtubule nucleation factor (TPX2), transcript variant X1	MODEL	mRNA
KIAA1324	XM_011541825.1	-6.037100	0.0360086	PREDICTED: Homo sapiens KIAA1324 (KIAA1324), transcript variant X1	MODEL	mRNA
LOC105372462	XR_919753.2	-5.409798	0.0360086	PREDICTED: Homo sapiens uncharacterized LOC105372462 (LOC105372462), transcript variant X2	MODEL	ncRNA
C12orf57	NM_001301834.1	-5.784194	0.0417056	chromosome 12 open reading frame 57 (C12orf57), transcript variant 2	REVIEWED	mRNA
MIR100HG	NR_137195.1	-6.471956	0.0417056	mir-100-let-7a-2-mir-125b-1 cluster host gene (MIR100HG), transcript variant 22	REVIEWED	long non-coding RNA
NFX1	XR_001746314.1	-6.104268	0.0417056	PREDICTED: Homo sapiens nuclear transcription factor, X-box binding 1 (NFX1), transcript variant X5	MODEL	misc_RNA
LINC01896	XR_935681.2	-5.985569	0.0417056	PREDICTED: Homo sapiens uncharacterized LOC645321 (LOC645321), transcript variant X3	MODEL	ncRNA
LINC01896	XR_952162.2	-5.985569	0.0417056	PREDICTED: Homo sapiens uncharacterized LOC645321 (LOC645321), transcript variant X3	MODEL	ncRNA
LINC01896	XR_952517.2	-5.985569	0.0417056	PREDICTED: Homo sapiens uncharacterized LOC645321 (LOC645321), transcript variant X3	MODEL	ncRNA
LOC105369351	XR_950214.2	-6.154897	0.0419230	PREDICTED: Homo sapiens uncharacterized LOC105369351 (LOC105369351), transcript variant X2	MODEL	ncRNA
SFXN5	XM_005264648.1	-6.738681	0.0441793	PREDICTED: Homo sapiens sideroflexin 5 (SFXN5), transcript variant X23	MODEL	mRNA
SELE	NM_000450.2	-4.639474	0.0449577	selectin E (SELE)	REVIEWED	mRNA
GADD45B	NM_015675.3	-3.501461	0.0449577	growth arrest and DNA damage inducible beta (GADD45B)	REVIEWED	mRNA
MTX3	XM_017009440.1	-5.907310	0.0449577	PREDICTED: Homo sapiens metaxin 3 (MTX3), transcript variant X2	MODEL	mRNA

Developmental Control - Autism

Child

This table shows the 22 most significant genes differentially expressed between control child samples and autism child samples AND different between control child samples and fetal samples.

genes_dev_control_autism_child <- signifChildCA %>%
  filter(target_id %in% devChildCA) %>%
  mutate(gene = text_spec(symbol, link = paste0('http://www.genecards.org/cgi-bin/carddisp.pl?gene=', symbol)),
         transcript = target_id) %>%
  dplyr::select(c('gene', 'transcript', 'log2FoldChange', 'padj', 'description', 'status', 'molecule_type'))
genes_dev_control_autism_child %>%
  kable() %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

gene	transcript	log2FoldChange	padj	description	status	molecule_type
H2AFY	NM_001040158.1	-8.025760	0.0028433	H2A histone family member Y (H2AFY), transcript variant 4	REVIEWED	mRNA
ZNF195	XM_011520352.2	6.694985	0.0030546	PREDICTED: Homo sapiens zinc finger protein 195 (ZNF195), transcript variant X7	MODEL	mRNA
SUSD5	XM_017006137.1	6.711799	0.0035603	PREDICTED: Homo sapiens sushi domain containing 5 (SUSD5), transcript variant X2	MODEL	mRNA
ACVR1C	NM_145259.2	8.014360	0.0039174	activin A receptor type 1C (ACVR1C), transcript variant 1	VALIDATED	mRNA
BUB1B	NM_001211.5	3.315151	0.0063885	BUB1 mitotic checkpoint serine/threonine kinase B (BUB1B)	REVIEWED	mRNA
ELAVL2	XM_011517776.2	5.712591	0.0069766	PREDICTED: Homo sapiens ELAV like neuron-specific RNA binding protein 2 (ELAVL2), transcript variant X1	MODEL	mRNA
ZFP64	XM_005260449.1	-7.105792	0.0084906	PREDICTED: Homo sapiens ZFP64 zinc finger protein (ZFP64), transcript variant X6	MODEL	mRNA
RGL1	XM_017000756.1	6.326049	0.0155666	PREDICTED: Homo sapiens ral guanine nucleotide dissociation stimulator like 1 (RGL1), transcript variant X3	MODEL	mRNA
ZNF160	NM_001322136.1	-6.555408	0.0156208	zinc finger protein 160 (ZNF160), transcript variant 12	REVIEWED	mRNA
CHD3	XM_017024064.1	7.552134	0.0156208	PREDICTED: Homo sapiens chromodomain helicase DNA binding protein 3 (CHD3), transcript variant X9	MODEL	mRNA
EXOC1	NM_018261.3	-6.547517	0.0162899	exocyst complex component 1 (EXOC1), transcript variant 1	REVIEWED	mRNA
TTR	NM_000371.3	7.428806	0.0207366	transthyretin (TTR)	REVIEWED	mRNA
PELI3	NM_001243136.1	-8.116381	0.0207366	pellino E3 ubiquitin protein ligase family member 3 (PELI3), transcript variant 4	REVIEWED	mRNA
LRRC75A-AS1	NR_027159.1	-5.821114	0.0207366	LRRC75A antisense RNA 1 (LRRC75A-AS1), transcript variant 3	PREDICTED	long non-coding RNA
PLPP5	XM_017013905.1	5.051513	0.0207366	PREDICTED: Homo sapiens phospholipid phosphatase 5 (PLPP5), transcript variant X10	MODEL	mRNA
NEUROD4	NM_021191.2	5.695382	0.0261895	neuronal differentiation 4 (NEUROD4)	VALIDATED	mRNA
MCM7	NM_001278595.1	-6.148432	0.0341248	minichromosome maintenance complex component 7 (MCM7), transcript variant 3	REVIEWED	mRNA
AP2S1	NM_001301081.1	-6.879410	0.0369278	adaptor related protein complex 2 sigma 1 subunit (AP2S1), transcript variant 5	REVIEWED	mRNA
RAB11FIP1	NM_025151.4	2.165864	0.0369278	RAB11 family interacting protein 1 (RAB11FIP1), transcript variant 1	VALIDATED	mRNA
SFRP1	NM_003012.4	2.112847	0.0385669	secreted frizzled related protein 1 (SFRP1)	REVIEWED	mRNA
A4GALT	XM_005261647.2	-6.759707	0.0428226	PREDICTED: Homo sapiens alpha 1,4-galactosyltransferase (A4GALT), transcript variant X4	MODEL	mRNA
TAGAP	NM_152133.2	5.808768	0.0497265	T-cell activation RhoGTPase activating protein (TAGAP), transcript variant 1	REVIEWED	mRNA

Adult

This table shows the 40 most significant genes differentially expressed between control adult samples and autism adult samples AND different between control adult samples and fetal samples.

genes_dev_control_autism_adult <- signifAdultCA %>%
  filter(target_id %in% devAdultCA) %>%
  mutate(gene = text_spec(symbol, link = paste0('http://www.genecards.org/cgi-bin/carddisp.pl?gene=', symbol)),
         transcript = target_id) %>%
  dplyr::select(c('gene', 'transcript', 'log2FoldChange', 'padj', 'description', 'status', 'molecule_type'))
genes_dev_control_autism_adult %>%
  kable() %>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

gene	transcript	log2FoldChange	padj	description	status	molecule_type
FHOD3	XM_005258355.1	-8.217616	0.0000000	PREDICTED: Homo sapiens formin homology 2 domain containing 3 (FHOD3), transcript variant X13	MODEL	mRNA
LOC107986119	XR_001750192.1	7.763706	0.0000087	PREDICTED: Homo sapiens uncharacterized LOC107986119 (LOC107986119), transcript variant X1	MODEL	ncRNA
DLGAP1	NM_001242764.1	-7.603438	0.0000505	DLG associated protein 1 (DLGAP1), transcript variant 6	VALIDATED	mRNA
AFF2	NM_001170628.1	-5.376843	0.0011518	AF4/FMR2 family member 2 (AFF2), transcript variant 6	REVIEWED	mRNA
CSE1L	NR_045796.1	-7.724335	0.0018763	chromosome segregation 1 like (CSE1L), transcript variant 3	REVIEWED	non-coding RNA
PRKCE	XM_006712050.3	-7.724396	0.0019385	PREDICTED: Homo sapiens protein kinase C epsilon (PRKCE), transcript variant X17	MODEL	mRNA
LOC107985374	XR_001735766.1	6.458495	0.0025999	PREDICTED: Homo sapiens uncharacterized LOC107985374 (LOC107985374)	MODEL	ncRNA
CHD3	XM_017024066.1	7.378109	0.0028590	PREDICTED: Homo sapiens chromodomain helicase DNA binding protein 3 (CHD3), transcript variant X11	MODEL	mRNA
ZNF821	XM_017023411.1	-6.516153	0.0028638	PREDICTED: Homo sapiens zinc finger protein 821 (ZNF821), transcript variant X4	MODEL	mRNA
OGT	XM_017029907.1	-7.653633	0.0034868	PREDICTED: Homo sapiens O-linked N-acetylglucosamine (GlcNAc) transferase (OGT), transcript variant X1	MODEL	mRNA
LOC107986119	XR_001740863.1	-7.333702	0.0036826	PREDICTED: Homo sapiens uncharacterized LOC107986119 (LOC107986119), transcript variant X1	MODEL	ncRNA
ANO1	NM_018043.5	5.930273	0.0047391	anoctamin 1 (ANO1), transcript variant 1	VALIDATED	mRNA
GRIA1	XM_017009393.1	-6.981561	0.0047391	PREDICTED: Homo sapiens glutamate ionotropic receptor AMPA type subunit 1 (GRIA1), transcript variant X3	MODEL	mRNA
EIF3C	XM_017023814.1	7.513897	0.0047391	PREDICTED: Homo sapiens eukaryotic translation initiation factor 3 subunit C (EIF3C), transcript variant X1	MODEL	mRNA
PASK	XM_011510835.1	-7.389713	0.0060558	PREDICTED: Homo sapiens PAS domain containing serine/threonine kinase (PASK), transcript variant X14	MODEL	mRNA
ATAD2	XM_011516996.2	-2.870322	0.0084250	PREDICTED: Homo sapiens ATPase family, AAA domain containing 2 (ATAD2), transcript variant X3	MODEL	mRNA
SCAF11	XM_017020221.1	6.422783	0.0100051	PREDICTED: Homo sapiens SR-related CTD associated factor 11 (SCAF11), transcript variant X8	MODEL	mRNA
WNK2	XM_011518936.2	-6.311372	0.0103810	PREDICTED: Homo sapiens WNK lysine deficient protein kinase 2 (WNK2), transcript variant X22	MODEL	mRNA
TMEM268	XM_017014429.1	5.870467	0.0103810	PREDICTED: Homo sapiens chromosome 9 open reading frame 91 (C9orf91), transcript variant X3	MODEL	mRNA
LSM6	XR_001741100.1	6.039595	0.0103810	PREDICTED: Homo sapiens LSM6 homolog, U6 small nuclear RNA and mRNA degradation associated (LSM6), transcript variant X3	MODEL	misc_RNA
SETD2	NM_001349370.1	-7.886412	0.0116732	SET domain containing 2 (SETD2), transcript variant 2	REVIEWED	mRNA
CHD3	XM_017024064.1	7.747000	0.0117789	PREDICTED: Homo sapiens chromodomain helicase DNA binding protein 3 (CHD3), transcript variant X9	MODEL	mRNA
ACVR1C	NM_001111033.1	-7.267053	0.0134341	activin A receptor type 1C (ACVR1C), transcript variant 4	VALIDATED	mRNA
UBA1	XM_017029780.1	-5.980508	0.0134341	PREDICTED: Homo sapiens ubiquitin like modifier activating enzyme 1 (UBA1), transcript variant X6	MODEL	mRNA
CGN	XM_005245365.4	7.766203	0.0136224	PREDICTED: Homo sapiens cingulin (CGN), transcript variant X1	MODEL	mRNA
TAB2	NM_001292035.2	-4.660715	0.0146400	TGF-beta activated kinase 1/MAP3K7 binding protein 2 (TAB2), transcript variant 4	REVIEWED	mRNA
PITPNM3	NM_001165966.1	-7.579690	0.0170745	PITPNM family member 3 (PITPNM3), transcript variant 2	REVIEWED	mRNA
TNFRSF10D	NM_003840.4	-2.259592	0.0323857	TNF receptor superfamily member 10d (TNFRSF10D)	REVIEWED	mRNA
SPAST	NM_014946.3	-4.304759	0.0360086	spastin (SPAST), transcript variant 1	REVIEWED	mRNA
TPX2	XM_011528697.2	5.036863	0.0360086	PREDICTED: Homo sapiens TPX2, microtubule nucleation factor (TPX2), transcript variant X1	MODEL	mRNA
KIAA1324	XM_011541825.1	-6.037100	0.0360086	PREDICTED: Homo sapiens KIAA1324 (KIAA1324), transcript variant X1	MODEL	mRNA
LOC105372462	XR_919753.2	-5.409798	0.0360086	PREDICTED: Homo sapiens uncharacterized LOC105372462 (LOC105372462), transcript variant X2	MODEL	ncRNA
C12orf57	NM_001301834.1	-5.784194	0.0417056	chromosome 12 open reading frame 57 (C12orf57), transcript variant 2	REVIEWED	mRNA
MIR100HG	NR_137195.1	-6.471956	0.0417056	mir-100-let-7a-2-mir-125b-1 cluster host gene (MIR100HG), transcript variant 22	REVIEWED	long non-coding RNA
NFX1	XR_001746314.1	-6.104268	0.0417056	PREDICTED: Homo sapiens nuclear transcription factor, X-box binding 1 (NFX1), transcript variant X5	MODEL	misc_RNA
LINC01896	XR_935681.2	-5.985569	0.0417056	PREDICTED: Homo sapiens uncharacterized LOC645321 (LOC645321), transcript variant X3	MODEL	ncRNA
LINC01896	XR_952162.2	-5.985569	0.0417056	PREDICTED: Homo sapiens uncharacterized LOC645321 (LOC645321), transcript variant X3	MODEL	ncRNA
LINC01896	XR_952517.2	-5.985569	0.0417056	PREDICTED: Homo sapiens uncharacterized LOC645321 (LOC645321), transcript variant X3	MODEL	ncRNA
SFXN5	XM_005264648.1	-6.738681	0.0441793	PREDICTED: Homo sapiens sideroflexin 5 (SFXN5), transcript variant X23	MODEL	mRNA
GADD45B	NM_015675.3	-3.501461	0.0449577	growth arrest and DNA damage inducible beta (GADD45B)	REVIEWED	mRNA
MTX3	XM_017009440.1	-5.907310	0.0449577	PREDICTED: Homo sapiens metaxin 3 (MTX3), transcript variant X2	MODEL	mRNA
NCKAP5	XM_017003980.1	-6.068193	0.0466438	PREDICTED: Homo sapiens NCK associated protein 5 (NCKAP5), transcript variant X18	MODEL	mRNA
FHL2	NM_001318896.1	-5.358883	0.0495433	four and a half LIM domains 2 (FHL2), transcript variant 8	REVIEWED	mRNA
TREM2	XM_006715116.3	5.479870	0.0495433	PREDICTED: Homo sapiens triggering receptor expressed on myeloid cells 2 (TREM2), transcript variant X1	MODEL	mRNA

The list of “developmental” genes that are differentially expressed in Autism vs Control are saved in DESeq2_glist_child.txt and DESeq2_glist_adult.txt (Entrez GeneIDs), in DESeq2_slist_child.txt and DESeq2_slist_adult.txt (gene symbols) while the list of “developmental” transcripts are saved in DESeq2_tlist_child.txt and DESeq2_tlist_adult.txt.

Save gene lists

unlink('DESeq2_tlist_child.txt')
unlink('DESeq2_tlist_adult.txt')
unlink('DESeq2_glist_child.txt')
unlink('DESeq2_glist_adult.txt')
unlink('DESeq2_slist_child.txt')
unlink('DESeq2_slist_adult.txt')
lapply(devChildCA[order(devChildCA)], write, 'DESeq2_tlist_child.txt', append=TRUE)
lapply(devAdultCA[order(devAdultCA)], write, 'DESeq2_tlist_adult.txt', append=TRUE)
lapply(devChildCAg[order(devChildCAg)], write, 'DESeq2_glist_child.txt', append=TRUE)
lapply(devAdultCAg[order(devAdultCAg)], write, 'DESeq2_glist_adult.txt', append=TRUE)
lapply(devChildCAs[order(devChildCAs)], write, 'DESeq2_slist_child.txt', append=TRUE)
lapply(devAdultCAs[order(devAdultCAs)], write, 'DESeq2_slist_adult.txt', append=TRUE)
save(file='DESeq2_tlist.RData', devChildCA, devAdultCA)
save(file='DESeq2_glist.RData', devChildCAg, devAdultCAg, devChildCAs, devAdultCAs)

GO analysis

Perform GO analysis using the goseq package, which normalizes GO term counts by transcript length.

refseq.lengths <- read_tsv('/share/db/refseq/human/refseq.108.rna.lengths', col_names = c('target_id', 'length'))
ttg <- ttg %>% full_join(refseq.lengths, by = 'target_id')
gene_lengths <- ttg %>%
  dplyr::select(gene_id,length) %>%
  filter(!is.na(gene_id)) %>%
  group_by(gene_id) %>%
  summarise(median_length = median(length)) %>%
  dplyr::select(gene_id, median_length) %>%
  spread(gene_id, median_length) %>%
  as.list() %>%
  unlist()

genes <- rep(0, length(unique(na.omit(ttg$gene_id))))
names(genes) <- na.omit(unique(ttg$gene_id))
genes[as.character(unique(signifChildCA$gene_id))] <- 1

pwf <- nullp(DEgenes = genes, genome = NULL, id = NULL, bias.data = gene_lengths)

GO.wall <- goseq(pwf, "hg19", "knownGene")
t1 <- GO.wall %>% filter(over_represented_pvalue < 0.05) %>% group_by(ontology) %>% summarise(n=n())
t2 <- GO.wall %>% filter(ontology == 'BP' & over_represented_pvalue < 0.05 & grepl('(neur[oa])|nerv', term)) %>%
  mutate(category = paste0('[',category,'](http://amigo.geneontology.org/amigo/term/',category,')')) %>%
  dplyr::select(c('category', 'term', 'numDEInCat', 'numInCat', 'over_represented_pvalue'))
t3 <- GO.wall %>% filter(ontology == 'MF' & over_represented_pvalue < 0.05) %>% head(n=50) %>% 
  mutate(category = paste0('[',category,'](http://amigo.geneontology.org/amigo/term/',category,')')) %>%
  dplyr::select(c('category', 'term', 'numDEInCat', 'numInCat', 'over_represented_pvalue'))
t4 <- GO.wall %>% filter(ontology == 'CC' & over_represented_pvalue < 0.05) %>% head(n=50) %>% 
  mutate(category = paste0('[',category,'](http://amigo.geneontology.org/amigo/term/',category,')')) %>%
  dplyr::select(c('category', 'term', 'numDEInCat', 'numInCat', 'over_represented_pvalue'))

Summary

Number of significant categories in different ontologies

kable(t1, format = 'markdown')

ontology	n
BP	232
CC	38
MF	42

Biological process

Summary of biological process terms related to neuronal development

kable(t2, format = 'markdown')

category	term	numDEInCat	numInCat	over_represented_pvalue
GO:0014034	neural crest cell fate commitment	1	3	0.0058216
GO:1904956	regulation of midbrain dopaminergic neuron differentiation	1	6	0.0105133
GO:1904338	regulation of dopaminergic neuron differentiation	1	11	0.0199172
GO:0061351	neural precursor cell proliferation	2	129	0.0211787
GO:0014029	neural crest formation	1	12	0.0216980
GO:0001764	neuron migration	2	136	0.0232999
GO:0090179	planar cell polarity pathway involved in neural tube closure	1	13	0.0234826
GO:0090178	regulation of establishment of planar polarity involved in neural tube closure	1	14	0.0245459
GO:0090177	establishment of planar polarity involved in neural tube closure	1	15	0.0260297
GO:1904948	midbrain dopaminergic neuron differentiation	1	19	0.0327061
GO:0097150	neuronal stem cell population maintenance	1	20	0.0332975

Molecular function

Summary of the 50th most significant “molecular functions” terms

kable(t3, format = 'markdown')

category	term	numDEInCat	numInCat	over_represented_pvalue
GO:0050512	lactosylceramide 4-alpha-galactosyltransferase activity	1	1	0.0011524
GO:0004003	ATP-dependent DNA helicase activity	2	35	0.0017298
GO:0001011	transcription factor activity, sequence-specific DNA binding, RNA polymerase recruiting	1	1	0.0019134
GO:0001087	transcription factor activity, TFIIB-class binding	1	1	0.0019134
GO:0003678	DNA helicase activity	2	50	0.0034513
GO:0038100	nodal binding	1	2	0.0038874
GO:0008321	Ral guanyl-nucleotide exchange factor activity	1	3	0.0052964
GO:0001093	TFIIB-class transcription factor binding	1	3	0.0057301
GO:0004674	protein serine/threonine kinase activity	4	445	0.0068994
GO:0016361	activin receptor activity, type I	1	4	0.0069912
GO:0008094	DNA-dependent ATPase activity	2	76	0.0077956
GO:0001083	transcription factor activity, RNA polymerase II basal transcription factor binding	1	5	0.0086088
GO:0010385	double-stranded methylated DNA binding	1	5	0.0090954
GO:0016773	phosphotransferase activity, alcohol group as acceptor	5	774	0.0098344
GO:0046935	1-phosphatidylinositol-3-kinase regulator activity	1	6	0.0108505
GO:0070324	thyroid hormone binding	1	6	0.0110543
GO:0017002	activin-activated receptor activity	1	7	0.0123594
GO:0035014	phosphatidylinositol 3-kinase regulator activity	1	7	0.0127756
GO:0008026	ATP-dependent helicase activity	2	101	0.0131991
GO:0070035	purine NTP-dependent helicase activity	2	101	0.0131991
GO:0016301	kinase activity	5	842	0.0135987
GO:0000182	rDNA binding	1	10	0.0166580
GO:0015643	toxic substance binding	1	11	0.0186086
GO:0003677	DNA binding	9	2451	0.0191138
GO:0035615	clathrin adaptor activity	1	11	0.0197205
GO:0098748	endocytic adaptor activity	1	11	0.0197205
GO:0001091	RNA polymerase II basal transcription factor binding	1	11	0.0198866
GO:0005068	transmembrane receptor protein tyrosine kinase adaptor activity	1	11	0.0199631
GO:0004672	protein kinase activity	4	640	0.0233940
GO:0008195	phosphatidate phosphatase activity	1	13	0.0244220
GO:0016772	transferase activity, transferring phosphorus-containing groups	5	983	0.0250600
GO:0017049	GTP-Rho binding	1	17	0.0260892
GO:0004386	helicase activity	2	150	0.0274406
GO:0004675	transmembrane receptor protein serine/threonine kinase activity	1	16	0.0294806
GO:0030674	protein binding, bridging	2	162	0.0318956
GO:0005540	hyaluronic acid binding	1	21	0.0360257
GO:0060090	binding, bridging	2	180	0.0386527
GO:0015248	sterol transporter activity	1	28	0.0457912
GO:0005539	glycosaminoglycan binding	2	200	0.0465436
GO:0019207	kinase regulator activity	2	201	0.0468330
GO:0043425	bHLH transcription factor binding	1	27	0.0477401
GO:0046982	protein heterodimerization activity	3	477	0.0487781

Cellular component

Summary of the 50th most significant “cellular component” terms

kable(t4, format = 'markdown')

category	term	numDEInCat	numInCat	over_represented_pvalue
GO:0048179	activin receptor complex	1	3	0.0054682
GO:0098592	cytoplasmic side of apical plasma membrane	1	3	0.0058180
GO:0001740	Barr body	1	5	0.0081692
GO:0044454	nuclear chromosome part	4	493	0.0098116
GO:0000805	X chromosome	1	7	0.0105260
GO:0001739	sex chromatin	1	6	0.0111291
GO:0000228	nuclear chromosome	4	527	0.0123948
GO:0000778	condensed nuclear chromosome kinetochore	1	9	0.0144316
GO:0042555	MCM complex	1	11	0.0189376
GO:0000784	nuclear chromosome, telomeric region	2	131	0.0209653
GO:0000940	condensed chromosome outer kinetochore	1	12	0.0218776
GO:0098687	chromosomal region	3	349	0.0218971
GO:0030122	AP-2 adaptor complex	1	13	0.0226759
GO:0030128	clathrin coat of endocytic vesicle	1	14	0.0238303
GO:0036020	endolysosome membrane	1	14	0.0242340
GO:0016581	NuRD complex	1	17	0.0279754
GO:0090545	CHD-type complex	1	17	0.0279754
GO:0035098	ESC/E(Z) complex	1	18	0.0306395
GO:0000780	condensed nuclear chromosome, centromeric region	1	19	0.0309061
GO:0000145	exocyst	1	19	0.0312876
GO:0030132	clathrin coat of coated pit	1	18	0.0318834
GO:0005942	phosphatidylinositol 3-kinase complex	1	20	0.0322205
GO:0005721	pericentric heterochromatin	1	20	0.0326199
GO:0000242	pericentriolar material	1	20	0.0331107
GO:0000781	chromosome, telomeric region	2	167	0.0333106
GO:0036019	endolysosome	1	20	0.0342002
GO:0005622	intracellular	29	14144	0.0343890
GO:0030666	endocytic vesicle membrane	2	167	0.0353067
GO:0005680	anaphase-promoting complex	1	21	0.0356366
GO:1990234	transferase complex	4	738	0.0374282
GO:0030125	clathrin vesicle coat	1	24	0.0413874
GO:0000775	chromosome, centromeric region	2	188	0.0414829
GO:0032993	protein-DNA complex	2	192	0.0423233
GO:0000803	sex chromosome	1	26	0.0436667
GO:0034451	centriolar satellite	1	26	0.0452129
GO:0000785	chromatin	3	470	0.0463529
GO:0005829	cytosol	13	4749	0.0466718
GO:0051233	spindle midzone	1	28	0.0469542

Heatmap of DE genes

Child

myttg <- ttg %>% filter(target_id %in% devChildCA) %>% dplyr::select(c('target_id', 'symbol'))
symbols <- myttg$symbol
names(symbols) <- myttg$target_id
vst <- varianceStabilizingTransformation(ddsWald, blind=FALSE)
obs <- assay(vst) %>%
 as.data.frame() %>%
 rownames_to_column('target_id') %>%
 filter(target_id %in% devChildCA) %>%
 gather(colnames(assay(ddsWald)), key = 'sample', value = 'est_counts') %>%
 spread(target_id, est_counts) %>%
 remove_rownames() %>%
 column_to_rownames(var="sample") %>%
 as.matrix()
colnames(obs) <- symbols[colnames(obs)]
labels <- metadata1 %>% dplyr::select(c('sample','age')) %>% arrange(sample) %>%
 remove_rownames() %>% column_to_rownames(var='sample')
Rowv <- obs %>%
  dist(method="euclidean") %>%
  hclust(method = "complete") %>%
  as.dendrogram %>%
  # rotate(c(6:7,1:5)) %>%
  color_branches(k = 3) %>%
  set("branches_lwd", 4)
obs[which(obs == 0)] <- 1
par(mar=c(5,4,4,2)+0.1)
heatmap.2(obs, Rowv = Rowv, trace="none", margins=c(5,9), srtCol = 45, 
          symm = F, symkey = F, symbreaks = F, scale="none")

Adult

myttg <- ttg %>% filter(target_id %in% devAdultCA) %>% dplyr::select(c('target_id', 'symbol'))
symbols <- myttg$symbol
names(symbols) <- myttg$target_id
vst <- varianceStabilizingTransformation(ddsWald, blind=FALSE)
obs <- assay(vst) %>%
 as.data.frame() %>%
 rownames_to_column('target_id') %>%
 filter(target_id %in% devAdultCA) %>%
 gather(colnames(assay(ddsWald)), key = 'sample', value = 'est_counts') %>%
 spread(target_id, est_counts) %>%
 remove_rownames() %>%
 column_to_rownames(var="sample") %>%
 as.matrix()
colnames(obs) <- symbols[colnames(obs)]
labels <- metadata1 %>% dplyr::select(c('sample','age')) %>% arrange(sample) %>%
 remove_rownames() %>% column_to_rownames(var='sample')
Rowv <- obs %>%
  dist(method="euclidean") %>%
  hclust(method = "complete") %>%
  as.dendrogram %>%
  # rotate(c(6:7,1:5)) %>%
  color_branches(k = 3) %>%
  set("branches_lwd", 4)
obs[which(obs == 0)] <- 1
par(mar=c(5,4,4,2)+0.1)
heatmap.2(obs, Rowv = Rowv, trace="none", margins=c(5,9), srtCol = 45,
          symm = F, symkey = F, symbreaks = F, scale="none")

Correlation plot for all genes

This correlation plot uses all the transcripts annotated as PROVISIONAL, REVIEWED or VALIDATED in RefSeq (the high confidence categories)

cor_tlist <- ttg %>% filter(status %in% c('PROVISIONAL', 'REVIEWED', 'VALIDATED')) %>% .[['target_id']]
obs <- counts(ddsWald, normalized=TRUE) %>%
  as.data.frame() %>%
  rownames_to_column('target_id') %>%
  filter(target_id %in% cor_tlist) %>%
  gather(colnames(assay(ddsWald)), key = 'sample', value = 'est_counts') %>% 
  full_join(metadata1,  by = 'sample') %>%
  dplyr::select(c('grp','target_id','est_counts')) %>%
  group_by(grp, target_id) %>%
  summarise(est_counts = mean(est_counts)) %>%
  spread(target_id, est_counts) %>%
  remove_rownames() %>%
  column_to_rownames(var="grp") %>%
  t() %>%
  as.tibble() %>%
  filter_at(vars(matches('Child|Young|Middle|Old')), any_vars(. >= 0.5))
colnames(obs) <- metadata1 %>% 
  dplyr::select(grp) %>% 
  group_by(grp) %>% 
  summarise(group_count=n()) %>% 
  mutate(group = paste0(grp,'(',group_count,')')) %>%
  .[['group']]
colpal=brewer.pal(n=8, name="RdBu")
corrplot(cor(obs),
         method="circle",
         type="lower",
         col=colpal,
         cl.lim=c(0,1),
         number.cex = 0.6,
         addCoef.col = "black",
         tl.col="black",
         tl.cex = 0.75,
         tl.srt=45,
         diag=FALSE)

Correlation plot for DE genes

This correlation plot uses the ‘significant’ transcripts, which are DE between Fetal and Control samples and also between Autism and Control samples.

obs <- counts(ddsWald, normalized=TRUE) %>%
  as.data.frame() %>%
  rownames_to_column('target_id') %>%
  filter(target_id %in% devChildCA) %>%
  gather(colnames(assay(ddsWald)), key = 'sample', value = 'est_counts') %>% 
  full_join(metadata1,  by = 'sample') %>%
  dplyr::select(c('grp','target_id','est_counts')) %>%
  group_by(grp, target_id) %>%
  summarise(est_counts = mean(est_counts)) %>%
  spread(target_id, est_counts) %>%
  remove_rownames() %>%
  column_to_rownames(var="grp") %>%
  t() %>%
  as.tibble() %>%
  filter_at(vars(matches('Child|Young|Middle|Old')), any_vars(. >= 0.5))
colnames(obs) <- metadata1 %>% 
  dplyr::select(grp) %>% 
  group_by(grp) %>% 
  summarise(group_count=n()) %>% 
  mutate(group = paste0(grp,'(',group_count,')')) %>%
  .[['group']]
colpal=brewer.pal(n=8, name="RdBu")
corrplot(cor(obs),
         method="circle",
         type="lower",
         col=colpal,
         #cl.lim=c(0,1),
         number.cex = 0.6,
         addCoef.col = "black",
         tl.col="black",
         tl.cex = 0.75,
         tl.srt=45,
         diag=FALSE)