tcga: TCGA data with somatic non-synonymous single nucleotide...

In c7rishi/variantprobs: Computing Probabilities of Gene Variants

Description

A subset of the publicly available TCGA whole exome sequencing data containing non-synonymous SNV mutations. More specifically, the subset of the TCGA data with Variant_Type == "SNP" and Variant_Classification c("Missense_Mutation", "Nonsense_Mutation", "Nonstop_Mutation").

Usage

tcga

Format

A dataset with 1,991,488 rows and 4 columns:

Hugo_Symbol

the gene label

Variant

the variant label. Obtained by concatenating the columns labeled 'Hugo_Symbol', 'Chromosome', 'Start_Position', 'Tumor_Seq_Allele1', and 'Tumor_Seq_Allele2' in the original TCGA data

patient_id

the patient (tumor) label. Obtained by extracting the first 16 characters of the column'Tumor_Sample_Barcode' in the original TCGA data

Cancer_Code

the cancer category associated with the tumor. Obtained by matching the TSS code (the 6th and 7th characters of 'Tumor_Sample_Barcode' column of the original TCGA data) with publicly available TSS study names.

MS

dominant Mutation Signature of the tumor. Obtained using Alexandrov 2013 mutation signature calling algorithm. Contains 7 categories Non-hypermutated, Smoking (4), Other, APOBEC (2, 13), UV (7), POLE (10), MMR (6, 15, 20, 26), with the numbers within parenthesis indicating the Signature number in COSMIC-30 list. The category Non-hypermutated corresponds to tumors with total number of mutations <= 500.

Source

https://gdc.cancer.gov/about-data/publications/mc3-2017

References

Alexandrov, Ludmil B., et al. "Signatures of mutational processes in human cancer." Nature 500.7463 (2013): 415-421.

c7rishi/variantprobs documentation built on June 23, 2020, 7:42 a.m.