VCFsToCatalogs: Create SBS, DBS and Indel catalogs from VCFs

In ICAMS: In-Depth Characterization and Analysis of Mutational Signatures ('ICAMS')

Create SBS, DBS and Indel catalogs from VCFs

Description

Create 3 SBS catalogs (96, 192, 1536), 3 DBS catalogs (78, 136, 144) and Indel catalog from the Mutect VCFs specified by files

Usage

VCFsToCatalogs(
  files,
  ref.genome,
  variant.caller = "unknown",
  num.of.cores = 1,
  trans.ranges = NULL,
  region = "unknown",
  names.of.VCFs = NULL,
  tumor.col.names = NA,
  filter.status = NULL,
  get.vaf.function = NULL,
  ...,
  max.vaf.diff = 0.02,
  return.annotated.vcfs = FALSE,
  suppress.discarded.variants.warnings = TRUE
)

Arguments

files	Character vector of file paths to the VCF files.
ref.genome	A ref.genome argument as described in ICAMS.
variant.caller	Name of the variant caller that produces the VCF, can be either "strelka", "mutect", "freebayes" or "unknown". This information is needed to calculate the VAFs (variant allele frequencies). If variant caller is "unknown"(default) and get.vaf.function is NULL, then VAF and read depth will be NAs. If variant caller is "mutect", do not merge SBSs into DBS.
num.of.cores	The number of cores to use. Not available on Windows unless num.of.cores = 1.
trans.ranges	Optional. If ref.genome specifies one of the BSgenome object BSgenome.Hsapiens.1000genomes.hs37d5 BSgenome.Hsapiens.UCSC.hg38 BSgenome.Mmusculus.UCSC.mm10 then the function will infer trans.ranges automatically. Otherwise, user will need to provide the necessary trans.ranges. Please refer to TranscriptRanges for more details. If is.null(trans.ranges) do not add transcript range information.
region	A character string designating a genomic region; see as.catalog and ICAMS.
names.of.VCFs	Optional. Character vector of names of the VCF files. The order of names in names.of.VCFs should match the order of VCF file paths in files. If NULL(default), this function will remove all of the path up to and including the last path separator (if any) in files and file paths without extensions (and the leading dot) will be used as the names of the VCF files.
tumor.col.names	Optional. Only applicable to Mutect VCFs. Character vector of column names in Mutect VCFs which contain the tumor sample information. The order of names in tumor.col.names should match the order of Mutect VCFs specified in files. If tumor.col.names is equal to NA(default), this function will use the 10th column in all the Mutect VCFs to calculate VAFs. See GetMutectVAF for more details.
filter.status	The status indicating a variant has passed all filters. An example would be "PASS". Variants which don't have the specified filter.status in the FILTER column in VCF will be removed. If NULL(default), no variants will be removed from the original VCF.
get.vaf.function	Optional. Only applicable when variant.caller is "unknown". Function to calculate VAF(variant allele frequency) and read depth information from original VCF. See GetMutectVAF as an example. If NULL(default) and variant.caller is "unknown", then VAF and read depth will be NAs.
...	Optional arguments to get.vaf.function.
max.vaf.diff	Not applicable if variant.caller = "mutect". The maximum difference of VAF, default value is 0.02. If the absolute difference of VAFs for adjacent SBSs is bigger than max.vaf.diff, then these adjacent SBSs are likely to be "merely" asynchronous single base mutations, opposed to a simultaneous doublet mutation or variants involving more than two consecutive bases.
return.annotated.vcfs	Logical. Whether to return the annotated VCFs with additional columns showing mutation class for each variant. Default is FALSE.
suppress.discarded.variants.warnings	Logical. Whether to suppress warning messages showing information about the discarded variants. Default is TRUE.

Details

This function calls VCFsToSBSCatalogs, VCFsToDBSCatalogs and VCFsToIDCatalogs

Value

A list containing the following objects:

• catSBS96, catSBS192, catSBS1536: Matrix of 3 SBS catalogs (one each for 96, 192, and 1536).

• catDBS78, catDBS136, catDBS144: Matrix of 3 DBS catalogs (one each for 78, 136, and 144).

• catID: Matrix of ID (small insertion and deletion) catalog.

• discarded.variants: Non-NULL only if there are variants that were excluded from the analysis. See the added extra column discarded.reason for more details.

• annotated.vcfs: Non-NULL only if return.annotated.vcfs = TRUE. A list of elements:

  • SBS: SBS VCF annotated by AnnotateSBSVCF with three new columns SBS96.class, SBS192.class and SBS1536.class showing the mutation class for each SBS variant.

  • DBS: DBS VCF annotated by AnnotateDBSVCF with three new columns DBS78.class, DBS136.class and DBS144.class showing the mutation class for each DBS variant.

  • ID: ID VCF annotated by AnnotateIDVCF with one new column ID.class showing the mutation class for each ID variant.

If trans.ranges is not provided by user and cannot be inferred by ICAMS, SBS 192 and DBS 144 catalog will not be generated. Each catalog has attributes added. See as.catalog for more details.

ID classification

See https://github.com/steverozen/ICAMS/blob/master/data-raw/PCAWG7_indel_classification_2021_09_03.xlsx for additional information on ID (small insertion and deletion) mutation classification.

See the documentation for Canonicalize1Del which first handles deletions in homopolymers, then handles deletions in simple repeats with longer repeat units, (e.g. CACACACA, see FindMaxRepeatDel), and if the deletion is not in a simple repeat, looks for microhomology (see FindDelMH).

See the code for unexported function CanonicalizeID and the functions it calls for handling of insertions.

Note

SBS 192 and DBS 144 catalogs include only mutations in transcribed regions. In ID (small insertion and deletion) catalogs, deletion repeat sizes range from 0 to 5+, but for plotting and end-user documentation deletion repeat sizes range from 1 to 6+.

Comments

To add or change attributes of the catalog, you can use function attr.
For example, attr(catalog, "abundance") <- custom.abundance.

Examples

file <- c(system.file("extdata/Mutect-vcf",
                      "Mutect.GRCh37.s1.vcf",
                      package = "ICAMS"))
if (requireNamespace("BSgenome.Hsapiens.1000genomes.hs37d5", quietly = TRUE)) {
  catalogs <- VCFsToCatalogs(file, ref.genome = "hg19",
                             variant.caller = "mutect", region = "genome")}

ICAMS documentation built on June 22, 2024, 6:47 p.m.