Cancer
A.Carolina Ledezma-Carrizalez
2022-08-01
R Markdown
This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.
Grouping and combining data.
In this part, I am going to try to reproduce the session from week 5 of the materials for an R course run by the Bioinformatics Core at the Cancer Research UK Cambridge Institute. The difference will be that the data used will be from Pediatric Brain Cancer https://www.cbioportal.org/study/clinicalData?id=brain_cptac_2020. With some functions of the dplyr package.
I’ll try to customize the plots using ggplot2. These dplyr and ggplot2 are core component packages within the tidyverse and both are loaded as part of the tidyverse.
library(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
## ✔ ggplot2 3.3.6 ✔ purrr 0.3.4
## ✔ tibble 3.1.8 ✔ dplyr 1.0.9
## ✔ tidyr 1.2.0 ✔ stringr 1.4.0
## ✔ readr 2.1.2 ✔ forcats 0.5.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()brain <- read_tsv("brain_cptac_2020_clinical_data.tsv")## Rows: 218 Columns: 63
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: "\t"
## chr (51): Study ID, Patient ID, Sample ID, Age Class, BRAF_RELA Status, BRAF...
## dbl (12): AGE, Age at Chemotherapy Start, Age at Chemotherapy Stop, Age at I...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.head(brain, n = 4)## # A tibble: 4 × 63
## Study …¹ Patie…² Sampl…³ AGE Age a…⁴ Age a…⁵ Age a…⁶ Age a…⁷ Age a…⁸ Age a…⁹
## <chr> <chr> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 brain_c… PT_CWD… 7316-1… 9 NA NA 3472 4008 NA NA
## 2 brain_c… PT_5FR… 7316-1… 11 NA NA 4234 6510 NA NA
## 3 brain_c… PT_RM5… 7316-1… 12 4630 4720 4519 6220 4556 4600
## 4 brain_c… PT_3X3… 7316-1… 30 11201 11242 11184 11881 11215 11247
## # … with 53 more variables: `Age at Specimen Diagnosis` <dbl>,
## # `Age Class` <chr>, `BRAF_RELA Status` <chr>, `BRAF Status` <chr>,
## # `BRAF Status2` <chr>, `Cancer Predispositions` <chr>, `Cancer Type` <chr>,
## # `Cancer Type Detailed` <chr>, Chemotherapy <chr>,
## # `Chemotherapy Agents` <chr>, `Chemotherapy Type` <chr>,
## # `Clinical Status at Collection Event` <chr>,
## # `Completed Radiation Dose` <chr>, `Completed Total Radiation Dose` <chr>, …
## # ℹ Use `colnames()` to see all variable namestable(brain$Chemotherapy)##
## No Not Reported Unavailable Yes
## 136 22 3 48summary(brain$`Mutation Count`)## Min. 1st Qu. Median Mean 3rd Qu. Max. NA's
## 3.00 13.00 27.00 44.94 59.50 587.00 18We can create a scatter plot to see if there is any relationship between the age of the patient at diagnosis of brain cancer and the Nottingham prognostic index (whatever that is).
plot(brain$`Age at Initial Diagnosis`, brain$`Nottingham prognostic index`,
xlab = "Age at diagnosis", ylab = "Nottingham prognostic index")## Warning: Unknown or uninitialised column: `Nottingham prognostic index`.
library(readr)
braincsv <- read.csv2("brain_cptac_2020_clinical_data.csv", header = TRUE, sep = ";", dec = ",",fill = TRUE)
head(braincsv, n = 4)## Study.ID Patient.ID Sample.ID AGE Age.at.Chemotherapy.Start
## 1 brain_cptac_2020 PT_CWD717Q0 7316-101 9 NA
## 2 brain_cptac_2020 PT_5FR2YA6E 7316-109 11 NA
## 3 brain_cptac_2020 PT_RM5S859Q 7316-111 12 4630
## 4 brain_cptac_2020 PT_3X3MF8ZD 7316-114 30 11201
## Age.at.Chemotherapy.Stop Age.at.Initial.Diagnosis
## 1 NA 3472
## 2 NA 4234
## 3 4720 4519
## 4 11242 11184
## Age.at.Last.Known.Clinical.Status Age.at.Radiation.Start
## 1 4008 NA
## 2 6510 NA
## 3 6220 4556
## 4 11881 11215
## Age.at.Radiation.Stop Age.at.Specimen.Diagnosis Age.Class BRAF_RELA.Status
## 1 NA 3472 [5-10) <NA>
## 2 NA 4234 [10-15) <NA>
## 3 4600 4519 [10-15) RELA Fusion
## 4 11247 11184 20+ <NA>
## BRAF.Status BRAF.Status2 Cancer.Predispositions CancerType
## 1 <NA> <NA> None documented Glioma
## 2 <NA> <NA> None documented Embryonal Tumor
## 3 RELA Fusion <NA> None documented CNS Cancer
## 4 <NA> <NA> Not Reported High-grade glioma/astrocytoma
## Cancer.Type.Detailed Chemotherapy
## 1 Ganglioglioma No
## 2 Medulloblastoma Yes
## 3 Ependymomal Tumor Yes
## 4 Pediatric High Grade Gliomas Yes
## Chemotherapy.Agents
## 1 <NA>
## 2 Cisplatin, Lomustine
## 3 Cisplatin, Cyclophosphamide, Etoposide, Vincristine
## 4 Temozolamide
## Chemotherapy.Type
## 1 <NA>
## 2 Treatment follows a protocol but subject is not enrolled
## 3 Treatment follows a protocol and subject is enrolled on a protocol
## 4 Treatment follows other standard of care not associated with a current or past protocol
## Clinical.Status.at.Collection.Event Completed.Radiation.Dose
## 1 Alive <NA>
## 2 Alive <NA>
## 3 Alive 5940 cGy
## 4 Alive 2340 cGy
## Completed.Total.Radiation.Dose CTNNB1.Status DFS.Months DFS.Status
## 1 <NA> <NA> 17 0:DiseaseFree
## 2 <NA> <NA> 47 1:Recurred/Progressed
## 3 5940 cGy CTNNB1 WT 55 0:DiseaseFree
## 4 4500 cGy <NA> 22 0:DiseaseFree
## Ependymoma_RELA.Status Ethnicity Extent.of.Tumor.Resection
## 1 <NA> Not Hispanic or Latino Gross/Near total resection
## 2 <NA> Not Hispanic or Latino <NA>
## 3 RELA Fusion Not Hispanic or Latino Gross/Near total resection
## 4 <NA> Not Hispanic or Latino Biopsy only
## External.Patient.ID
## 1 C23862
## 2 C28905
## 3 C30381
## 4 C30258
## Formulation
## 1 <NA>
## 2 Reduced cisplatin dose by 50% due to high frequency hearing loss and reduced CCNU dose by 25% due to persistent neutropenia.
## 3 Not Reported
## 4 Not Reported
## H3F3A_CTNNB1.Status HGG_H3F3A.Status
## 1 <NA> <NA>
## 2 <NA> <NA>
## 3 CTNNB1 WT <NA>
## 4 H3F3A WT H3F3A WT
## Initial.CNS.Tumor.Diagnosis.Related.to.OS
## 1 Ganglioglioma
## 2 Medulloblastoma
## 3 Ependymoma
## 4 High-grade glioma/astrocytoma
## Initial.Diagnosis.Type LGG_BRAF.Status
## 1 Initial CNS Tumor <NA>
## 2 Initial CNS Tumor <NA>
## 3 Initial CNS Tumor <NA>
## 4 Second Malignancy (first HGG record) <NA>
## Medical.Conditions
## 1 Seizure disorder
## 2 Other medical conditions NOS
## 3 Focal neurological deficit (cranial nerve palsies, motor deficits, sensory deficits),Seizure disorder
## 4 Focal neurological deficit (cranial nerve palsies, motor deficits, sensory deficits),Other medical conditions NOS
## Multiple.Cancer.Predispositions Multiple.Medical.Conditions
## 1 No No
## 2 No No
## 3 No Yes
## 4 <NA> Yes
## Multiple.Tumor.Locations Mutation.Count Oncotree.Code OS.Months OS.Status
## 1 No 69 GNG 17 0:LIVING
## 2 No 41 MBL 74 1:DECEASED
## 3 No 40 EPMT 55 0:LIVING
## 4 Yes 46 PHGG 22 1:DECEASED
## Protocol.and.Treatment.Arm Race Radiation
## 1 <NA> Black or African American No
## 2 CHP 455 White No
## 3 COG trial ACNS0831 White Yes
## 4 Not Reported White Yes
## Radiation.Site Radiation.Type
## 1 <NA> <NA>
## 2 <NA> <NA>
## 3 Focal/Tumor bed Protons
## 4 Craniospinal with focal boost Combination of Protons and Photons
## Sample.Annotation Number.of.Samples.Per.Patient Sample.Origin
## 1 Treatment naive 1 Initial CNS Tumor Surgery
## 2 Treatment naive 1 <NA>
## 3 Treatment naive 1 Initial CNS Tumor Surgery
## 4 Treatment naive 1 Second malignancy
## Sex Surgery Timing.Other.Notes Treatment Treatment.Changed
## 1 Male Yes <NA> New Diagnosis
## 2 Male No Not Reported Modified Treatment 12 Month Update
## 3 Female Yes Not Reported New Diagnosis
## 4 Female Yes Not Reported New Diagnosis
## Treatment.Status Tumor.Location.Condensed
## 1 Treatment naive Temporal Lobe
## 2 Treatment naive Cerebellar
## 3 Treatment naive Supratentorial
## 4 Treatment naive Midline
## Tumor.Tissue.Site
## 1 Temporal Lobe
## 2 Cerebellum/Posterior Fossa
## 3 Occipital Lobe
## 4 Brain Stem- Pons,Brain Stem-Medulla,Spinal Cord- Cervical,Spinal Cord- Thoracic,Suprasellar/Hypothalamic/Pituitary
## Tumor.Type Updated.Grade
## 1 Primary I
## 2 Primary IV
## 3 Primary III
## 4 Second Malignancy IV#library(ggplot2)
ggplot(data = braincsv) +
geom_point(mapping = aes(x =AGE , y = CancerType))
ggplot(data = braincsv) +
geom_point(mapping = aes(x = AGE, y = CancerType, colour = OS.Status))
ggplot(data = braincsv) +
geom_point(mapping = aes(x = AGE, y = CancerType, colour = "Nottingham_prognostic_index"))
ggplot(data = braincsv) +
geom_point(mapping = aes(x = AGE, y = CancerType, size = OS.Status))
ggplot(data = braincsv) +
geom_bar(mapping = aes(x = CancerType))+
theme(axis.text.x = element_text(angle = 15, size = 10))
ggplot(data = braincsv) +
geom_bar(mapping = aes(x = CancerType, fill = CancerType))+
theme(axis.text.x = element_text(angle = 15, size = 10))
ggplot(data = braincsv) +
geom_bar(mapping = aes(x = CancerType, fill = Sex))+
theme(axis.text.x = element_text(angle = 15, size = 10))
ggplot(data = braincsv, mapping = aes(x = CancerType, y = AGE)) +
geom_point(size = 2, alpha = 1) +
theme(axis.text.x = element_text(angle = 15, size = 10))+
geom_smooth()
ggplot(data = braincsv, mapping = aes(x = AGE, y = CancerType, colour = OS.Status)) +
geom_point(size = 2, alpha = 1) +
theme(axis.text.x = element_text(angle = 15, size = 10))
## Box plots
ggplot(data = braincsv, mapping = aes(x = AGE, y = CancerType)) +
geom_boxplot()
See geom_boxplot help to explain how the box and whiskers are
constructed and how it decides which points are outliers and should be
displayed as points.
ggplot(data = braincsv, mapping = aes(x = AGE, y = CancerType)) +
geom_boxplot() +
geom_point()
Here we can notice an atypical value, an age of 30 years in pediatric
brain cancer, I think it is outside the pediatric margin
Ideally, we’d like these points to be spread out a bit. The geom_point help points to geom_jitter as more suitable when one of the variables is categorical.
ggplot(data = braincsv, mapping = aes(x = AGE, y = CancerType)) +
geom_boxplot() +
geom_jitter(width = 0.3, size = 0.8, alpha = 0.25)
ggplot(data = braincsv, mapping = aes(x = AGE, y = CancerType, colour = OS.Status)) +
geom_boxplot()