maftools is a comprehensive toolkit for processing somatic variants from cohort-based cancer genomic studies. maftools offers over 80 functions to perform the most commonly required tasks in cancer genomics, using MAF as the only input file type.
#Install from Bioconductor repository
BiocManager::install("maftools")
#Install from GitHub repository
BiocManager::install("PoisonAlien/maftools")
A complete documentation of maftools using TCGA LAML as a case study can be found here.
maftools is extremely easy to use, starting with importing an MAF file along with the associated clinical data. Once the data is successfully imported, the resulting MAF object can be passed to various functions. Key applications include:
- Cohort summarization using oncoplots
- Identify co-occurring and mutually exclusive events
- Clinical enrichment analysis
- Detect cancer driver genes
- Infer tumor heterogeneity
- Analyze known cancer signaling pathways
- De-novo somatic signature analysis with NMF
- Compare two cohorts to identify differentially mutated genes
- Perform survival analysis and predict genesets associated with survival
- Drug-gene interactions
Besides the MAF files, maftools can handle sequencing alignment BAM files, copy number output from GISTIC and mosdepth. Please refer to the package documentation sections below to learn more.
- Generate personalized cancer report for known somatic hotspots
- Sample mismatch and relatedness analysis
- Copy number analysis with ASCAT and mosdepth
Moreover, analyzing all 33 TCGA cohorts along with the harmonized clinical data is a breeze.
- A single command tcgaLoad will import the desired TCGA cohort thereby avoiding costly time spent on data mining from public databases.
- Please refer to an associated software package TCGAmutations that provides ready to use
MAFobjects for 33 TCGA cohorts and 2427 cell line profiles from CCLE - along with relevant clinical information for all sequenced samples.
Mayakonda A, Lin DC, Assenov Y, Plass C, Koeffler HP. 2018. Maftools: efficient and comprehensive analysis of somatic variants in cancer. Genome Research. PMID: 30341162
| File Fomats | Data portals | Annotation tools |
|---|---|---|
| Mutation Annotation Format | TCGA | vcf2maf - for converting your VCF files to MAF |
| Variant Call Format | ICGC | annovar2maf - for converting annovar output files to MAF |
| ICGC Simple Somatic Mutation Format | Broad Firehose | bcftools csq - Rapid annotations of VCF files with variant consequences |
| cBioPortal | Annovar | |
| PeCan | Funcotator | |
| CIViC - Clinical interpretation of variants in cancer | ||
| DGIdb - Information on drug-gene interactions and the druggable genome |
Below are some more useful software packages for somatic variant analysis
- TRONCO - Repository of the TRanslational ONCOlogy library (R)
- dndscv - dN/dS methods to quantify selection in cancer and somatic evolution (R)
- cloneevol - Inferring and visualizing clonal evolution in multi-sample cancer sequencing (R)
- sigminer - Primarily for signature analysis and visualization in R. Supports
maftoolsoutput (R) - GenVisR - Primarily for visualization (R)
- comut - Primarily for visualization (Python)
- TCGAmutations - pre-compiled curated somatic mutations from TCGA cohorts (from Broad Firehose and TCGA MC3 Project) that can be loaded into
maftools(R) - somaticfreq - rapid genotyping of known somatic hotspot variants from the tumor BAM files. Generates a browsable/sharable HTML report. (C)
