inst/guides/general.guide.md

In steverozen/mSigAct.server: Shiny Server for Analyzing Mutational Signature Activities

Table of contents

• Introduction
• Mutational signature examples
• Background
• Workflow
• Creating mutational spectrum catalogs
• Estimating mutational signature activities
• Browser compatibility
• Code availability
• Getting help

Introduction to the mSigAct web server

The mSigAct web server provides methods for analyzing mutational signatures. “mSigAct” is an abbreviation of mutational Signature Activity. A mutational signature is the pattern of mutations generated by a particular mutational process. (At least, this is the ideal that we usually aim for.)

Mutational signature examples

Example single base substitution signature: mutations from CG to TG

These are “single base substitutions” (SBSs) caused by deamination of 5-methyl cytosine. We most often look at SBS mutations in the context of preceding and following bases. Also, by convention, we usually look a mutations from C or T to another base. We reverse complement mutations from G or A to another base. In the figure above we see mutational signature “SBS1”, which is the signature of 5-methyl cytosine deamination. It consists almost entirely of mutations from ACG to ATG, CCG to CTG, GCG to GTG and TCG to TTG. Signatures based on SBS mutations in the context of the preceding and following bases are called “SBS96 signatures”. A compendium with standard nomenclature is at the COSMIC Single Base Substitution web site.

We can also look at single base substitutions in the context of the preceding and following 2 bases, giving us “SBS1536 signatures”. We also sometimes look at signatures of single base substitution mutations in transcripts by distinguishing their transcribed (antisense) and untranscrbed (sense) strand, giving us “SBS192 signatures.”

Example insertion and deletion signature: moderately-sized deletions with microhomology

These are common in tumors with defective homologous-recombination-based DNA-damage repair, often BRCA1 or BRCA2-deficient tumors. The classification for insertions and deletions is detailed at https://www.synapse.org/#!Synapse:syn11801742.

Example doublet base substitution signature: CC to TT mutations

These are a kind of of “doublet base substitution”, (DBS) mutations that stem from ultraviolet-radiation-induced pyrimidine dimers. The classification of DBSs is detailed at https://www.synapse.org/#!Synapse:syn11801895

A mutational spectrum is usually an overlay of multple mutational signatures

A single tumor or tissue sample usually contains mutations from multiple mutational processes, each of which generates mutations that when combined constitute the mutational spectrum of the sample, as shown in this figure:

Reproduced from Alexandrov et al., 2020, under the Creative Commons Attribution 4.0 International License.

This breast cancer has an SBS96 spectrum that is an overlay of signatures SBS1, SBS2, SBS3, SBS5, and SBS13, with SBS3 (the SBS signature of defective homologous-recombination-based DNA repair) contributing the the bulk of the mutations. The breast cancer’s DBS spectrum has signatures DBS2, DBS4, DBS6, and DBS9. Its insertion and deletion (indel) spectrum has indel signatures ID1, ID6, and ID8. The latter two indel signatures are characteristic of tumors with defective homologous-recombination-based DNA repair, which is consistent with the large number of mutations due to SBS3.

For more extensive background, please see Alexandrov et al., 2020.

The a list and plots of the the currently known mutational signatures, please see https://cancer.sanger.ac.uk/cosmic/signatures/index.tt.

Overall workflow

Using the mSigAct web server to create mutational spectrum catalogs from variant call files and plot them

Please see the “Generate spectrum catalogs from VCFs” tab for more information and examples. The mSigAct web server can generate and plot mutational spectra from VCF files generated by the Strelka or Mutect variant callers.

Using the mSigAct web server to estimate mutational signature activities

By “mutational signature activities” we refer to the numbers of mutations generated by each mutational signature (or more precisely, generated by each mutational process). For the breast cancer example above, this would involve estimating which mutational signatures (processes) generated the somatic mutations and how many mutations each signature generated. This is in fact a challenging problem, and the purpose of this functionality in the mSigAct web server is to allow you to consider and examine all the evidence in this analysis. Please see the “Signature attribution” tab for more information and examples.

mSigAct web server browser compatibility

The mSigAct web server has been tested under the following browsers and operating systems

OS Version Chrome Edge Firefox Safari Linux   Ubuntu 20.04      86.0.4240.198    N/A     82.0.3     N/A MacOS Mojave 87.0.4280.88 87.0.664.60 84.0 14.0 Windows 10 87.0.4280.88 87.0.664.60 83.0 N/A

Code availability

The mSigAct web server code is available at github at https://github.com/steverozen/mSigAct.server and is released under the GPL3 open source license. Other code used by the web server includes mSigAct, ICAMS, ICAMSxtra, and PCAWG7, all released under the GPL3 open source license.

Getting help

To report issues, make suggestions, and request help post on the mSigAct webserver GitHub site or contact [e-mail].

steverozen/mSigAct.server documentation built on July 9, 2023, 4:52 a.m.