1 Summary

1.1 TCF7L2 family and subfamily

load("~/paper_elmer/ESCA_analysis_hg38/escc_vs_esad_hg38/hyper/ELMER_results_hyper.rda")
family <- ELMER:::getdata("TF.family")
family$TF7L2_HUMAN.H11MO.0.A

## [1] "LEF1"   "TAF1"   "TCF7"   "TCF7L1" "TCF7L2"

subfamily <- ELMER:::getdata("TF.subfamily")
subfamily$TF7L2_HUMAN.H11MO.0.A

## [1] "TCF7L2"

1.2 Enriched motifs and probes associated

# which is the enriched motif in the family
enriched.motif$TF7L2_HUMAN.H11MO.0.A

##  [1] "cg18645642" "cg07965110" "cg13784017" "cg11959316" "cg23172911"
##  [6] "cg17488500" "cg08529825" "cg04108706" "cg13420364" "cg17170262"
## [11] "cg06640374" "cg14181111" "cg21167643" "cg13420366" "cg10092394"
## [16] "cg08178956" "cg15153684" "cg13851334" "cg00470044" "cg08847775"
## [21] "cg15357118" "cg10560561" "cg12336777" "cg11554249" "cg16360836"
## [26] "cg20784693" "cg25029529" "cg17546376" "cg26681383" "cg18720486"
## [31] "cg11578055" "cg22173632" "cg13625875" "cg26306329" "cg03873281"
## [36] "cg08812108" "cg17450505" "cg12422539" "cg14151158" "cg01302136"
## [41] "cg23226631" "cg10784258" "cg04825872" "cg17285448" "cg02704552"
## [46] "cg17544672" "cg15822656" "cg16848937" "cg20411555" "cg01003666"
## [51] "cg11065621" "cg23625106" "cg07273304" "cg13575422" "cg14610066"
## [56] "cg03013917" "cg19749898" "cg04064032" "cg10687219" "cg20289045"
## [61] "cg23788418" "cg15689991" "cg01493728" "cg06255601" "cg14881470"
## [66] "cg17388689" "cg18836689" "cg14845609" "cg23551120" "cg07800658"
## [71] "cg17882867" "cg13594138" "cg18593236" "cg10330955" "cg09734761"
## [76] "cg06545761" "cg14008999" "cg18052778" "cg09545579" "cg07855465"
## [81] "cg01303222" "cg10237442" "cg18614510" "cg16071713" "cg05602642"
## [86] "cg12509665" "cg09049140" "cg27564716"

enriched.motif$TCF7_HUMAN.H11MO.0.A

##  [1] "cg07965110" "cg13784017" "cg11959316" "cg17488500" "cg08529825"
##  [6] "cg18113107" "cg13420364" "cg17170262" "cg06640374" "cg14181111"
## [11] "cg21167643" "cg13420366" "cg10092394" "cg21535106" "cg01273330"
## [16] "cg15153684" "cg13851334" "cg03687070" "cg00470044" "cg08847775"
## [21] "cg15357118" "cg23367358" "cg04330597" "cg16360836" "cg20784693"
## [26] "cg03217587" "cg17546376" "cg11666343" "cg08004073" "cg01377807"
## [31] "cg16217908" "cg15848095" "cg11578055" "cg11419262" "cg09852389"
## [36] "cg23971987" "cg18949666" "cg26728709" "cg19719207" "cg26306329"
## [41] "cg22078781" "cg08721802" "cg26678138" "cg05600609" "cg18259348"
## [46] "cg08812108" "cg17450505" "cg12422539" "cg14151158" "cg01302136"
## [51] "cg23226631" "cg04825872" "cg04757168" "cg00278517" "cg19929194"
## [56] "cg14434601" "cg17544672" "cg15822656" "cg16848937" "cg20411555"
## [61] "cg01003666" "cg14630106" "cg11065621" "cg13575422" "cg16319142"
## [66] "cg10476206" "cg03013917" "cg14255417" "cg13378013" "cg17425818"
## [71] "cg19749898" "cg04337594" "cg04064032" "cg26725502" "cg10687219"
## [76] "cg15085883" "cg09599740" "cg17388689" "cg18836689" "cg07800658"
## [81] "cg17882867" "cg13594138" "cg18593236" "cg01870976" "cg10330955"
## [86] "cg03752741" "cg09262171" "cg01540102" "cg14008999" "cg09780231"
## [91] "cg25293325" "cg23258611" "cg20921890" "cg16204205" "cg01303222"
## [96] "cg12067547" "cg18788725" "cg09049140"

length(union(enriched.motif$TF7L2_HUMAN.H11MO.0.A,
             enriched.motif$TCF7_HUMAN.H11MO.0.A
))

## [1] 136

length(intersect(enriched.motif$TF7L2_HUMAN.H11MO.0.A,
                 enriched.motif$TCF7_HUMAN.H11MO.0.A
))

## [1] 50

2 Narrow TCF7L2 peaks

colnames(peaks) <- c("chrom","Start","End","Name","Width","Score", "fold_enrichment","-log10(pValue)","-log10(qValue)","peak")
DT::datatable(peaks,options = list(scrollX = TRUE),filter = 'top')

peaks$chrom <- paste0("chr",peaks$chrom)
peaks.gr <- makeGRangesFromDataFrame(peaks,keep.extra.columns = T)
peaks.gr

## GRanges object with 28740 ranges and 6 metadata columns:
##           seqnames               ranges strand |            Name
##              <Rle>            <IRanges>  <Rle> |     <character>
##       [1]     chr1     [ 29198,  29441]      * |     OE33_peak_1
##       [2]     chr1     [713825, 714329]      * |     OE33_peak_2
##       [3]     chr1     [762669, 763049]      * |     OE33_peak_3
##       [4]     chr1     [839595, 840410]      * |     OE33_peak_4
##       [5]     chr1     [859188, 859486]      * |     OE33_peak_5
##       ...      ...                  ...    ... .             ...
##   [28736]     chrY [13460946, 13461195]      * | OE33_peak_28736
##   [28737]     chrY [13485442, 13485967]      * | OE33_peak_28737
##   [28738]     chrY [58827136, 58827588]      * | OE33_peak_28738
##   [28739]     chrY [58883594, 58884000]      * | OE33_peak_28739
##   [28740]     chrY [58905133, 58905450]      * | OE33_peak_28740
##                 Score fold_enrichment -log10(pValue) -log10(qValue)
##           <character>       <numeric>      <numeric>      <numeric>
##       [1]           .         4.15958        7.48895        4.98834
##       [2]           .         2.43645        4.85955        2.54250
##       [3]           .         5.53781       10.38868        7.74738
##       [4]           .         5.32820       14.39616       11.61202
##       [5]           .         6.29061       10.93385        8.26844
##       ...         ...             ...            ...            ...
##   [28736]           .         2.42299       10.26745        7.63054
##   [28737]           .         4.66213       19.55059       16.63674
##   [28738]           .        26.84596      104.76330      100.97492
##   [28739]           .         5.08153        7.61580        5.10279
##   [28740]           .        10.16306       21.14335       18.19341
##                peak
##           <integer>
##       [1]       122
##       [2]       215
##       [3]       198
##       [4]       461
##       [5]       116
##       ...       ...
##   [28736]       121
##   [28737]       177
##   [28738]       219
##   [28739]       185
##   [28740]       166
##   -------
##   seqinfo: 24 sequences from an unspecified genome; no seqlengths

ChIPseeker::covplot(peaks.gr, weightCol="fold_enrichment")

3 Comparing paired probes considering family classification

3.1 Get TCF7L2 probes target regions

pairs <- read_csv("~/paper_elmer/ESCA_analysis_hg38/escc_vs_esad_hg38/hyper/getTFtargets_genomic_coordinates_mapped_to_hg19.hyper.family.csv",
                  col_types = readr::cols())
TCF7L2_targets <- pairs[grep("TCF7L2",pairs$TF),]
DT::datatable(TCF7L2_targets,options = list(scrollX = TRUE),filter = 'top')

probes <- makeGRangesFromDataFrame(TCF7L2_targets,
                                   keep.extra.columns = T,
                                   seqnames.field = "probe_hg19_seqnames",
                                   start.field = "probe_hg19_start",
                                   end.field =  "probe_hg19_end",
                                   strand.field = "probe_hg19_strand"
)
probes <- unique(probes)
probes

## GRanges object with 136 ranges and 19 metadata columns:
##         seqnames               ranges strand |          GeneID       Probe
##            <Rle>            <IRanges>  <Rle> |     <character> <character>
##     [1]     chr6 [45441433, 45441434]      * | ENSG00000001561  cg04757168
##     [2]     chr4 [10447440, 10447441]      * | ENSG00000002587  cg11578055
##     [3]     chr4 [11652505, 11652506]      * | ENSG00000002587  cg23971987
##     [4]    chr11 [68174286, 68174287]      * | ENSG00000006534  cg10687219
##     [5]    chr19 [46385115, 46385116]      * | ENSG00000008438  cg12067547
##     ...      ...                  ...    ... .             ...         ...
##   [132]    chr15 [84168459, 84168460]      * | ENSG00000259767  cg03752741
##   [133]    chr15 [83790129, 83790130]      * | ENSG00000259767  cg18593236
##   [134]     chr7 [42303654, 42303655]      * | ENSG00000261019  cg14434601
##   [135]     chr7 [42195875, 42195876]      * | ENSG00000261019  cg16848937
##   [136]     chr7 [41721644, 41721645]      * | ENSG00000261019  cg17544672
##              Symbol       Sides        Raw.p          FDR distNearestTSS
##         <character> <character>    <numeric>    <numeric>      <integer>
##     [1]       ENPP4          R5 1.270122e-21 1.939775e-20         656295
##     [2]      HS3ST1          R7 1.237582e-10 3.681389e-10         982780
##     [3]      HS3ST1          L3 1.237582e-10 3.681389e-10         221115
##     [4]     ALDH3B1         L10 3.475069e-20 4.004857e-19         384974
##     [5]     PGLYRP1          R7 1.515588e-05 2.167515e-05         141207
##     ...         ...         ...          ...          ...            ...
##   [132]  AC022558.1          L8 0.0005811881 0.0006199670         475370
##   [133]  AC022558.1          L7 0.0005811881 0.0006199670          97040
##   [134]  AC010132.4          R5 0.0001512426 0.0001803295         638583
##   [135]  AC010132.4          R4 0.0001512426 0.0001803295         746362
##   [136]  AC010132.4          R8 0.0001512426 0.0001803295        1220592
##         DMC_analysis_pvalue DMC_analysis_escc_Minus_esad
##                   <numeric>                    <numeric>
##     [1]        1.540132e-27                    0.3424920
##     [2]        4.831871e-19                    0.3496876
##     [3]        1.074089e-22                    0.3277020
##     [4]        6.126029e-18                    0.3446468
##     [5]        2.397778e-48                    0.3446085
##     ...                 ...                          ...
##   [132]        7.073356e-19                    0.3031593
##   [133]        3.297960e-21                    0.3026496
##   [134]        2.581079e-27                    0.3200877
##   [135]        6.279606e-24                    0.3296102
##   [136]        1.009814e-19                    0.3272796
##         DMC_analysis_adjust.p external_gene_name original_ensembl_gene_id
##                     <numeric>        <character>              <character>
##     [1]          3.291965e-25              ENPP4        ENSG00000001561.6
##     [2]          3.257832e-17             HS3ST1        ENSG00000002587.8
##     [3]          1.202478e-20             HS3ST1        ENSG00000002587.8
##     [4]          3.579561e-16            ALDH3B1       ENSG00000006534.14
##     [5]          1.505816e-44            PGLYRP1        ENSG00000008438.4
##     ...                   ...                ...                      ...
##   [132]          4.668875e-17         AC022558.1        ENSG00000259767.1
##   [133]          3.001644e-19         AC022558.1        ENSG00000259767.1
##   [134]          5.446325e-25         AC010132.4        ENSG00000261019.1
##   [135]          8.298203e-22         AC010132.4        ENSG00000261019.1
##   [136]          7.469167e-18         AC010132.4        ENSG00000261019.1
##                                                                                                                                                                                        TF
##                                                                                                                                                                               <character>
##     [1]            FOXA3;FOXA2;FOXP4;FOXD2;FOXJ1;TCF7L2;NR1I2;PPARG;RORC;THRA;NR1H3;NR1H4;ZNF620;ZSCAN16;ZNF774;ZNF468;ZNF33A;ZNF816;ZNF619;ZNF765;ZNF799;ZNF763;ZNF823;ZNF124;SPDEF;ELF3
##     [2]                                           GATA6;GATA4;NR1I2;PPARG;RORC;THRA;NR1H3;NR1H4;TCF7L2;HNF1A;HNF1B;PDX1;CDX2;HOXB6;HOXB5;MNX1;EVX1;CDX1;HOXA13;HOXB9;SPDEF;ELF3;IRF8;SOX9
##     [3]                  GATA6;GATA4;NR1I2;PPARG;RORC;THRA;NR1H3;NR1H4;TCF7L2;SPDEF;ELF3;HNF1A;HNF1B;PDX1;CDX2;HOXB6;HOXB5;MNX1;EVX1;CDX1;HOXA13;HOXB9;FOXA3;FOXA2;FOXP4;FOXD2;FOXJ1;SOX9
##     [4]                                                                   HNF4A;HNF4G;NR2F2;NR2C1;FOXA3;FOXA2;FOXP4;FOXD2;FOXJ1;NR1I2;PPARG;RORC;THRA;NR1H3;NR1H4;TCF7L2;SPDEF;ELF3;NR3C2
##     [5] GATA6;GATA4;HNF1A;HNF1B;PDX1;CDX2;HOXB6;HOXB5;MNX1;EVX1;CDX1;HOXA13;HOXB9;SPDEF;ELF3;NR1I2;PPARG;RORC;THRA;NR1H3;NR1H4;TCF7L2;NFATC2;FOXA3;FOXA2;FOXP4;FOXD2;FOXJ1;ZSCAN16;ZNF816
##     ...                                                                                                                                                                               ...
##   [132]                                                                                HNF1A;HNF1B;PDX1;CDX2;HOXB6;HOXB5;MNX1;EVX1;CDX1;HOXA13;HOXB9;FOXA3;FOXA2;FOXP4;FOXD2;FOXJ1;TCF7L2
##   [133]                                              NR1I2;PPARG;RORC;THRA;NR1H3;NR1H4;HNF4A;HNF4G;NR2F2;NR2C1;TCF7L2;FOXA3;FOXA2;FOXP4;FOXD2;FOXJ1;NFATC2;SPDEF;ELF3;ZSCAN16;ZNF816;IRF8
##   [134]                                                                        GATA6;GATA4;NR1I2;PPARG;RORC;THRA;NR1H3;NR1H4;HNF4A;HNF4G;NR2F2;NR2C1;TCF7L2;FOXA3;FOXA2;FOXP4;FOXD2;FOXJ1
##   [135]                            HNF1A;HNF1B;PDX1;CDX2;HOXB6;HOXB5;MNX1;EVX1;CDX1;HOXA13;HOXB9;FOXA3;FOXA2;FOXP4;FOXD2;FOXJ1;TCF7L2;NR1I2;PPARG;RORC;THRA;NR1H3;NR1H4;NFATC2;SOX9;NR3C2
##   [136]                       HNF1A;HNF1B;PDX1;CDX2;HOXB6;HOXB5;MNX1;EVX1;CDX1;HOXA13;HOXB9;GATA6;GATA4;FOXA3;FOXA2;FOXP4;FOXD2;FOXJ1;NR1I2;PPARG;RORC;THRA;NR1H3;NR1H4;TCF7L2;SPDEF;ELF3
##         probe_hg19_width gene_hg19_seqnames gene_hg19_start gene_hg19_end
##                <integer>        <character>       <integer>     <integer>
##     [1]                2               chr6        46097730      46114436
##     [2]                2               chr4        11394774      11431389
##     [3]                2               chr4        11394774      11431389
##     [4]                2              chr11        67786065      67789293
##     [5]                2              chr19        46522411      46526323
##     ...              ...                ...             ...           ...
##   [132]                2              chr15        83690988      83693088
##   [133]                2              chr15        83690988      83693088
##   [134]                2               chr7        42940871      42942238
##   [135]                2               chr7        42940871      42942238
##   [136]                2               chr7        42940871      42942238
##         gene_hg19_width gene_hg19_strand
##               <integer>      <character>
##     [1]           16707                +
##     [2]           36616                -
##     [3]           36616                -
##     [4]            3229                +
##     [5]            3913                -
##     ...             ...              ...
##   [132]            2101                -
##   [133]            2101                -
##   [134]            1368                -
##   [135]            1368                -
##   [136]            1368                -
##   -------
##   seqinfo: 22 sequences from an unspecified genome; no seqlengths

3.2 Distance between probes and peaks

y <- distanceToNearest(probes,peaks.gr,ignore.strand=T)
qplot(mcols(y)$distance,geom="histogram") + ggthemes::theme_gdocs()

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

qplot(log(mcols(y)$distance + 1),geom="histogram") + ggthemes::theme_gdocs()

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

3.3 Results overlap

3.3.1 Overlap between probes (+-250bp) and peaks

probes_extended_500bp <- resize(unique(probes),width = 501,fix = "center")
probes_extended_500bp[,1:2]

## GRanges object with 136 ranges and 2 metadata columns:
##         seqnames               ranges strand |          GeneID       Probe
##            <Rle>            <IRanges>  <Rle> |     <character> <character>
##     [1]     chr6 [45441183, 45441683]      * | ENSG00000001561  cg04757168
##     [2]     chr4 [10447190, 10447690]      * | ENSG00000002587  cg11578055
##     [3]     chr4 [11652255, 11652755]      * | ENSG00000002587  cg23971987
##     [4]    chr11 [68174036, 68174536]      * | ENSG00000006534  cg10687219
##     [5]    chr19 [46384865, 46385365]      * | ENSG00000008438  cg12067547
##     ...      ...                  ...    ... .             ...         ...
##   [132]    chr15 [84168209, 84168709]      * | ENSG00000259767  cg03752741
##   [133]    chr15 [83789879, 83790379]      * | ENSG00000259767  cg18593236
##   [134]     chr7 [42303404, 42303904]      * | ENSG00000261019  cg14434601
##   [135]     chr7 [42195625, 42196125]      * | ENSG00000261019  cg16848937
##   [136]     chr7 [41721394, 41721894]      * | ENSG00000261019  cg17544672
##   -------
##   seqinfo: 22 sequences from an unspecified genome; no seqlengths

unique(width(probes_extended_500bp[,1:2]))

## [1] 501

hits <- suppressWarnings(findOverlaps(peaks.gr,probes_extended_500bp))
probes.hit <- subjectHits(hits)
length(unique(probes_extended_500bp[probes.hit])) # unique regions

## [1] 34

3.3.2 Overlap between peaks and distall probes (+-250bp)

distal.probes <- ELMER::get.feature.probe(feature = NULL,
                                          genome = "hg19", 
                                          met.platform = "450K")

## Downloading transcripts information. Using: Homo sapiens genes (GRCh37.p13)

## Returning distal probes: 163576

distal.probes.extended_500bp <- resize(unique(distal.probes),width = 501,fix = "center")

hits.distal <- suppressWarnings(findOverlaps(peaks.gr,distal.probes.extended_500bp))
distal.probes.hit <- subjectHits(hits.distal)

3.3.3 Regions overlapped

DT::datatable(tibble::as_data_frame(probes[subjectHits(hits)]),
              options = list(scrollX = TRUE),filter = 'top')

DT::datatable(tibble::as_data_frame(peaks.gr[queryHits(hits)]),
              options = list(scrollX = TRUE),filter = 'top')

3.3.4 Enrichement between peaks and all distal probes

x <- matrix(c(
  length(unique(probes[probes.hit])), # all linked probes with hits
  (length(unique(probes)) - length(unique(probes[probes.hit]))), # all linked probes with no hits
  length(unique(distal.probes.hit)), # all distal probes with  hits (4216)
  (length(distal.probes)) - length(unique(distal.probes.hit))), # all distal probes with no hits
  nrow = 2,
  dimnames = list(c("Overlap", "No overlap"),
                  c("Unique Linked probes", "Distal probes"))
)
x <- t(x)
x

##                      Overlap No overlap
## Unique Linked probes      34        102
## Distal probes           6406     157170

fisher.test(x)

## 
##  Fisher's Exact Test for Count Data
## 
## data:  x
## p-value < 2.2e-16
## alternative hypothesis: true odds ratio is not equal to 1
## 95 percent confidence interval:
##   5.372469 12.173156
## sample estimates:
## odds ratio 
##   8.179121

3.3.5 Enrichement between peaks and distal probes with signature for family members

motifs.probes <- ELMER:::getdata("Probes.motif.hg38.450K")

probes.family <- union(names(motifs.probes[motifs.probes[,"TCF7_HUMAN.H11MO.0.A"],"TCF7_HUMAN.H11MO.0.A"]),
                       names(motifs.probes[motifs.probes[,"TF7L2_HUMAN.H11MO.0.A"],"TF7L2_HUMAN.H11MO.0.A"])
)
distal.probes.family.motif <- distal.probes.extended_500bp[names(distal.probes) %in% probes.family]
hits.distal.family.motif <- suppressWarnings(findOverlaps(peaks.gr,distal.probes.family.motif))
distal.probes.family.motif.hit <- subjectHits(hits.distal.family.motif)

x <- matrix(c(
  length(unique(probes[probes.hit])), # all linked probes with hits
  (length(unique(probes)) - length(unique(probes[probes.hit]))), # all linked probes with no hits
  length(unique(distal.probes.family.motif.hit)), # all distal probes with  hits (4216)
  (length(distal.probes.family.motif)) - length(unique(distal.probes.family.motif.hit))), # all distal probes with no hits
  nrow = 2,
  dimnames = list(c("Overlap", "No overlap"),
                  c("Unique Linked probes", "Distal probes with family motif signature"))
)
x <- t(x)
x

##                                           Overlap No overlap
## Unique Linked probes                           34        102
## Distal probes with family motif signature    1212      15311

fisher.test(x)

## 
##  Fisher's Exact Test for Count Data
## 
## data:  x
## p-value = 2.442e-10
## alternative hypothesis: true odds ratio is not equal to 1
## 95 percent confidence interval:
##  2.755502 6.293503
## sample estimates:
## odds ratio 
##   4.210511

3.3.6 Enrichement between peaks and distal probes without signature for family members

motifs.probes <- ELMER:::getdata("Probes.motif.hg38.450K")

probes.family <- union(names(motifs.probes[motifs.probes[,"TCF7_HUMAN.H11MO.0.A"],"TCF7_HUMAN.H11MO.0.A"]),
                       names(motifs.probes[motifs.probes[,"TF7L2_HUMAN.H11MO.0.A"],"TF7L2_HUMAN.H11MO.0.A"])
)
distal.probes.family.nomotif <- distal.probes.extended_500bp[!names(distal.probes) %in% probes.family]
hits.distal.family.nomotif <- suppressWarnings(findOverlaps(peaks.gr,distal.probes.family.nomotif))
distal.probes.family.nomotif.hit <- subjectHits(hits.distal.family.nomotif)

x <- matrix(c(
  length(unique(probes[probes.hit])), # all linked probes with hits
  (length(unique(probes)) - length(unique(probes[probes.hit]))), # all linked probes with no hits
  length(unique(distal.probes.family.nomotif.hit)), # all distal probes with  hits (4216)
  (length(distal.probes.family.nomotif)) - length(unique(distal.probes.family.nomotif.hit))), # all distal probes with no hits
  nrow = 2,
  dimnames = list(c("Overlap", "No overlap"),
                  c("Unique Linked probes", "Distal probes without family motif signature"))
)
x <- t(x)
x

##                                              Overlap No overlap
## Unique Linked probes                              34        102
## Distal probes without family motif signature    5194     141859

fisher.test(x)

## 
##  Fisher's Exact Test for Count Data
## 
## data:  x
## p-value < 2.2e-16
## alternative hypothesis: true odds ratio is not equal to 1
## 95 percent confidence interval:
##   5.978474 13.554812
## sample estimates:
## odds ratio 
##   9.103863

3.4 Using Dechen peak files

library(readr)
OE33_TCF7L2_peaks_genes <- read_csv("~/Downloads/OE33-TCF7L2_peaks_genes.csv")

## Parsed with column specification:
## cols(
##   .default = col_character(),
##   Start = col_integer(),
##   End = col_integer(),
##   `Peak Score` = col_integer()
## )

## See spec(...) for full column specifications.

peaks.gr.rep2 <- makeGRangesFromDataFrame(OE33_TCF7L2_peaks_genes,keep.extra.columns = T)
hits <- suppressWarnings(findOverlaps(peaks.gr.rep2,probes_extended_500bp))
probes.hit <- subjectHits(hits)
length(unique(probes_extended_500bp[probes.hit])) # unique regions

## [1] 28

4 Comparing paired probes considering subfamily classification

pairs <- read_csv("~/paper_elmer/ESCA_analysis_hg38/escc_vs_esad_hg38/hyper/getTFtargets_genomic_coordinates_mapped_to_hg19.hyper.subfamily.csv",
                  col_types = readr::cols())
TCF7L2_targets <- pairs[grep("TCF7L2",pairs$TF),]
probes <- makeGRangesFromDataFrame(TCF7L2_targets,
                                   keep.extra.columns = T,
                                   seqnames.field = "probe_hg19_seqnames",
                                   start.field = "probe_hg19_start",
                                   end.field =  "probe_hg19_end",
                                   strand.field = "probe_hg19_strand"
)
probes <- unique(probes)
probes

## GRanges object with 88 ranges and 19 metadata columns:
##        seqnames                 ranges strand |          GeneID
##           <Rle>              <IRanges>  <Rle> |     <character>
##    [1]     chr4   [10447440, 10447441]      * | ENSG00000002587
##    [2]    chr11   [68174286, 68174287]      * | ENSG00000006534
##    [3]     chr3   [54700753, 54700754]      * | ENSG00000016391
##    [4]    chr14   [69708392, 69708393]      * | ENSG00000029364
##    [5]     chrX   [19011128, 19011129]      * | ENSG00000044446
##    ...      ...                    ...    ... .             ...
##   [84]     chr2 [145210222, 145210223]      * | ENSG00000257226
##   [85]    chr15 [ 71704440,  71704441]      * | ENSG00000259470
##   [86]    chr15 [ 83790129,  83790130]      * | ENSG00000259767
##   [87]     chr7 [ 42195875,  42195876]      * | ENSG00000261019
##   [88]     chr7 [ 41721644,  41721645]      * | ENSG00000261019
##              Probe      Symbol       Sides        Raw.p          FDR
##        <character> <character> <character>    <numeric>    <numeric>
##    [1]  cg11578055      HS3ST1          R7 1.237582e-10 3.681389e-10
##    [2]  cg10687219     ALDH3B1         L10 3.475069e-20 4.004857e-19
##    [3]  cg26681383        CHDH          L9 3.222326e-25 1.424268e-23
##    [4]  cg18836689     SLC39A9          R6 4.628200e-04 5.017466e-04
##    [5]  cg09049140       PHKA2          L1 4.356173e-04 4.746561e-04
##    ...         ...         ...         ...          ...          ...
##   [84]  cg12336777  AC079584.2          L5 2.538522e-04 2.907691e-04
##   [85]  cg17882867      KRT8P9          L9 7.753461e-08 1.564434e-07
##   [86]  cg18593236  AC022558.1          L7 5.811881e-04 6.199670e-04
##   [87]  cg16848937  AC010132.4          R4 1.512426e-04 1.803295e-04
##   [88]  cg17544672  AC010132.4          R8 1.512426e-04 1.803295e-04
##        distNearestTSS DMC_analysis_pvalue DMC_analysis_escc_Minus_esad
##             <integer>           <numeric>                    <numeric>
##    [1]         982780        4.831871e-19                    0.3496876
##    [2]         384974        6.126029e-18                    0.3446468
##    [3]         820335        2.055364e-20                    0.3578636
##    [4]         156338        1.476132e-30                    0.3106146
##    [5]           8411        1.950920e-34                    0.4063213
##    ...            ...                 ...                          ...
##   [84]         676581        5.069337e-18                    0.3687613
##   [85]         553387        1.337372e-43                    0.4510772
##   [86]          97040        3.297960e-21                    0.3026496
##   [87]         746362        6.279606e-24                    0.3296102
##   [88]        1220592        1.009814e-19                    0.3272796
##        DMC_analysis_adjust.p external_gene_name original_ensembl_gene_id
##                    <numeric>        <character>              <character>
##    [1]          3.257832e-17             HS3ST1        ENSG00000002587.8
##    [2]          3.579561e-16            ALDH3B1       ENSG00000006534.14
##    [3]          1.684670e-18               CHDH        ENSG00000016391.9
##    [4]          4.646150e-28            SLC39A9       ENSG00000029364.10
##    [5]          1.054464e-31              PHKA2       ENSG00000044446.10
##    ...                   ...                ...                      ...
##   [84]          2.989934e-16         AC079584.2        ENSG00000257226.1
##   [85]          3.458314e-40             KRT8P9        ENSG00000259470.2
##   [86]          3.001644e-19         AC022558.1        ENSG00000259767.1
##   [87]          8.298203e-22         AC010132.4        ENSG00000261019.1
##   [88]          7.469167e-18         AC010132.4        ENSG00000261019.1
##                                                                      TF
##                                                             <character>
##    [1]                                         GATA6;GATA4;PPARG;TCF7L2
##    [2]                             HNF4A;HNF4G;NR1H3;NR1H4;TCF7L2;NR3C2
##    [3]                                  NR1H3;NR1H4;PDX1;TCF7L2;ZSCAN16
##    [4]          HNF4A;HNF4G;ELF3;FOXD2;TCF7L2;NR1H3;NR1H4;NR2F2;ZSCAN16
##    [5]                                         FOXD2;FOXA3;FOXA2;TCF7L2
##    ...                                                              ...
##   [84]          GATA6;GATA4;FOXA3;FOXA2;FOXD2;ELF3;TCF7L2;ZSCAN16;NR3C2
##   [85]   GATA6;GATA4;ELF3;MECOM;CDX2;CDX1;TCF7L2;NR3C2;NR2F2;HOXB6;EVX1
##   [86]                                       PPARG;TCF7L2;NR2F2;ZSCAN16
##   [87] HNF1A;HNF1B;FOXA3;FOXA2;FOXD2;CDX2;CDX1;TCF7L2;NR1H3;NR1H4;NR3C2
##   [88]   HNF1A;HNF1B;GATA6;GATA4;NR1H3;NR1H4;PDX1;ELF3;TCF7L2;CDX2;CDX1
##        probe_hg19_width gene_hg19_seqnames gene_hg19_start gene_hg19_end
##               <integer>        <character>       <integer>     <integer>
##    [1]                2               chr4        11394774      11431389
##    [2]                2              chr11        67786065      67789293
##    [3]                2               chr3        53846362      53880417
##    [4]                2              chr14        69864732      69929105
##    [5]                2               chrX        18910418      19002716
##    ...              ...                ...             ...           ...
##   [84]                2               chr2       144524189     144533642
##   [85]                2              chr15        71151053      71152501
##   [86]                2              chr15        83690988      83693088
##   [87]                2               chr7        42940871      42942238
##   [88]                2               chr7        42940871      42942238
##        gene_hg19_width gene_hg19_strand
##              <integer>      <character>
##    [1]           36616                -
##    [2]            3229                +
##    [3]           34056                -
##    [4]           64374                +
##    [5]           92299                -
##    ...             ...              ...
##   [84]            9454                -
##   [85]            1449                +
##   [86]            2101                -
##   [87]            1368                -
##   [88]            1368                -
##   -------
##   seqinfo: 21 sequences from an unspecified genome; no seqlengths

probes_extended_500bp <- resize(unique(probes),width = 501,fix = "center")
probes_extended_500bp[,1:2]

## GRanges object with 88 ranges and 2 metadata columns:
##        seqnames                 ranges strand |          GeneID
##           <Rle>              <IRanges>  <Rle> |     <character>
##    [1]     chr4   [10447190, 10447690]      * | ENSG00000002587
##    [2]    chr11   [68174036, 68174536]      * | ENSG00000006534
##    [3]     chr3   [54700503, 54701003]      * | ENSG00000016391
##    [4]    chr14   [69708142, 69708642]      * | ENSG00000029364
##    [5]     chrX   [19010878, 19011378]      * | ENSG00000044446
##    ...      ...                    ...    ... .             ...
##   [84]     chr2 [145209972, 145210472]      * | ENSG00000257226
##   [85]    chr15 [ 71704190,  71704690]      * | ENSG00000259470
##   [86]    chr15 [ 83789879,  83790379]      * | ENSG00000259767
##   [87]     chr7 [ 42195625,  42196125]      * | ENSG00000261019
##   [88]     chr7 [ 41721394,  41721894]      * | ENSG00000261019
##              Probe
##        <character>
##    [1]  cg11578055
##    [2]  cg10687219
##    [3]  cg26681383
##    [4]  cg18836689
##    [5]  cg09049140
##    ...         ...
##   [84]  cg12336777
##   [85]  cg17882867
##   [86]  cg18593236
##   [87]  cg16848937
##   [88]  cg17544672
##   -------
##   seqinfo: 21 sequences from an unspecified genome; no seqlengths

unique(width(probes_extended_500bp[,1:2]))

## [1] 501

hits <- suppressWarnings(findOverlaps(peaks.gr,probes_extended_500bp))
probes.hit <- subjectHits(hits)
length(unique(probes_extended_500bp[probes.hit])) # unique regions

## [1] 28

4.1 Enrichement between peaks and distal probes

x <- matrix(c(
  length(unique(probes[probes.hit])), # all linked probes with hits
  (length(unique(probes)) - length(unique(probes[probes.hit]))), # all linked probes with no hits
  length(unique(distal.probes.hit)), # all distal probes with  hits (4216)
  (length(distal.probes)) - length(unique(distal.probes.hit))), # all distal probes with no hits
  nrow = 2,
  dimnames = list(c("Overlap", "No overlap"),
                  c("Unique Linked probes", "Distal probes"))
)
x <- t(x)
x

##                      Overlap No overlap
## Unique Linked probes      28         60
## Distal probes           6406     157170

fisher.test(x)

## 
##  Fisher's Exact Test for Count Data
## 
## data:  x
## p-value < 2.2e-16
## alternative hypothesis: true odds ratio is not equal to 1
## 95 percent confidence interval:
##   7.032183 18.221937
## sample estimates:
## odds ratio 
##   11.44814

4.2 Enrichement between peaks and distal probes with signature for subfamily members

motifs.probes <- ELMER:::getdata("Probes.motif.hg38.450K")

probes.subfamily <-  names(motifs.probes[motifs.probes[,"TF7L2_HUMAN.H11MO.0.A"],
                                         "TF7L2_HUMAN.H11MO.0.A"])


distal.probes.subfamily.motif <- distal.probes.extended_500bp[names(distal.probes.extended_500bp) %in% probes.subfamily]
hits.distal.subfamily.motif <- suppressWarnings(findOverlaps(peaks.gr,distal.probes.subfamily.motif))
distal.probes.subfamily.motif.hit <- subjectHits(hits.distal.subfamily.motif)

x <- matrix(c(
  length(unique(probes[probes.hit])), # all linked probes with hits
  (length(unique(probes)) - length(unique(probes[probes.hit]))), # all linked probes with no hits
  length(unique(distal.probes.subfamily.motif.hit)), # all distal probes with  hits (4216)
  (length(distal.probes.subfamily.motif)) - length(unique(distal.probes.subfamily.motif.hit))), # all distal probes with no hits
  nrow = 2,
  dimnames = list(c("Overlap", "No overlap"),
                  c("Unique Linked probes", "Distal probes with subfamily motif signature"))
)
x <- t(x)
x

##                                              Overlap No overlap
## Unique Linked probes                              28         60
## Distal probes with subfamily motif signature     942       9571

fisher.test(x)

## 
##  Fisher's Exact Test for Count Data
## 
## data:  x
## p-value = 1.938e-09
## alternative hypothesis: true odds ratio is not equal to 1
## 95 percent confidence interval:
##  2.898532 7.583840
## sample estimates:
## odds ratio 
##   4.740295

4.3 Enrichement between peaks and distal probes without signature for subfamily members

probes.subfamily <-  names(motifs.probes[motifs.probes[,"TF7L2_HUMAN.H11MO.0.A"],
                                         "TF7L2_HUMAN.H11MO.0.A"])

distal.probes.subfamily.nomotif <- distal.probes.extended_500bp[!names(distal.probes) %in% probes.subfamily]
hits.distal.subfamily.nomotif <- suppressWarnings(findOverlaps(peaks.gr,distal.probes.subfamily.nomotif))
distal.probes.subfamily.nomotif.hit <- subjectHits(hits.distal.subfamily.nomotif)

x <- matrix(c(
  length(unique(probes[probes.hit])), # all linked probes with hits
  (length(unique(probes)) - length(unique(probes[probes.hit]))), # all linked probes with no hits
  length(unique(distal.probes.subfamily.nomotif.hit)), # all distal probes with  hits (4216)
  (length(distal.probes.subfamily.nomotif)) - length(unique(distal.probes.subfamily.nomotif.hit))), # all distal probes with no hits
  nrow = 2,
  dimnames = list(c("Overlap", "No overlap"),
                  c("Unique Linked probes", "Distal probes without subfamily motif signature"))
)
x <- t(x)
x

##                                                 Overlap No overlap
## Unique Linked probes                                 28         60
## Distal probes without subfamily motif signature    5464     147599

fisher.test(x)

## 
##  Fisher's Exact Test for Count Data
## 
## data:  x
## p-value < 2.2e-16
## alternative hypothesis: true odds ratio is not equal to 1
## 95 percent confidence interval:
##   7.741489 20.063776
## sample estimates:
## odds ratio 
##   12.60256

4.4 Using Dechen peak files

library(readr)
OE33_TCF7L2_peaks_genes <- read_csv("~/Downloads/OE33-TCF7L2_peaks_genes.csv")

## Parsed with column specification:
## cols(
##   .default = col_character(),
##   Start = col_integer(),
##   End = col_integer(),
##   `Peak Score` = col_integer()
## )

## See spec(...) for full column specifications.

peaks.gr.rep2 <- makeGRangesFromDataFrame(OE33_TCF7L2_peaks_genes,keep.extra.columns = T)
hits <- suppressWarnings(findOverlaps(peaks.gr.rep2,probes_extended_500bp))
probes.hit <- subjectHits(hits)
length(unique(probes_extended_500bp[probes.hit])) # unique regions

## [1] 23

Evaluation enrichment of TCF7L2 targets

1 Summary

1.1 TCF7L2 family and subfamily

1.2 Enriched motifs and probes associated

2 Narrow TCF7L2 peaks

3 Comparing paired probes considering family classification

3.1 Get TCF7L2 probes target regions

3.2 Distance between probes and peaks

3.3 Results overlap

3.3.1 Overlap between probes (+-250bp) and peaks

3.3.2 Overlap between peaks and distall probes (+-250bp)

3.3.3 Regions overlapped

3.3.4 Enrichement between peaks and all distal probes

3.3.5 Enrichement between peaks and distal probes with signature for family members

3.3.6 Enrichement between peaks and distal probes without signature for family members

3.4 Using Dechen peak files

4 Comparing paired probes considering subfamily classification

4.1 Enrichement between peaks and distal probes

4.2 Enrichement between peaks and distal probes with signature for subfamily members

4.3 Enrichement between peaks and distal probes without signature for subfamily members

4.4 Using Dechen peak files