Skip to content

Latest commit

 

History

History
168 lines (113 loc) · 9.62 KB

README.md

File metadata and controls

168 lines (113 loc) · 9.62 KB
Raw

SOPRANO: Selection On PRotein ANnotated regiOns

------------ Please use our new github version of the tool -> https://github.com/instituteofcancerresearch/SOPRANO/ ------------

------------ Even better use our webserver! https://software.icr.ac.uk/app/soprano ---------------------------------------------

SOPRANO method was developed to quantify selection in specific regions of the genome (Zapata et al, in revision). It calculates ON- and OFF-target dN/dS using a set of annotated somatic point mutations and a transcript coordinates file.

The set of mutations (missense/truncating) and their respective functional annotation using ensemblVEP. Mutations can be filtered a priori by the user (i.e. by predicted binding affinity or by expression status).

The set of ensembl transcript coordinates where selection will be estimated. ON-dN/dS is the value calculated inside the coordinates provided using a 192-trinucleotide correction signature obtained "on-the-fly" from the input mutation file. Alternatively, the user can provide a pre-calculated trinucleotide mutation frequency file. Importantly, ON dN/dS and OFF dN/dS (the portion outside the coordinates provided) will be calculated only using transcripts defined in this file.

Installation

Create a directory and clone the repository

mkdir tools
cd tools

git clone https://github.com/luisgls/SOPRANO.git

Install dependencies

Copy a local version and edit the master script to run locally (run_localSSBselection_v4.sh)

cp run_localSSBselection_v4.sh run_localSSBselection_v4_local.sh

  • Specify the base directory of the installation BASEDIR=/the/home/directory/SOPRANO/

  • Specify a tmp directory TMP=/my/tmp/

  • Specify the reference genome sequence fasta file (e.g. hg19.fasta) FASTA=/my/directory/to/hg19.fasta

  • Specify the reference genome length file (e.g. hg19.genome) GENOME=/my/directory/to/hg19.genome

Input file : Mutations

  • Mutations annotated with VEP using default output format (https://www.ensembl.org/info/docs/tools/vep/vep_formats.html) The input file is the standard output of variant effect predictor using the following command line (by providing to vep the ensembl default input file format) perl variant_effect_predictor.pl -i input -o input.annotated --cache --all_refseq --assembly GRCh37 --pick --symbol --no_stats --fasta genome.fasta If you want to filter putative germline variants use the option --plugin ExAC when running VEP. It is important that you restrict your analysis to the list of ensemb transcripts observed in data folder, be aware of updates on the EnsemblID from previous versions.

  • VCF file annotated with VEP using VCF format To convert a VCF annotated file to the input format for SOPRANO, the user can run vep-annotation-reporter from (https://vatools.readthedocs.io/en/latest/vep_annotation_reporter.html) using the following command:

vep-annotation-reporter -o OUTPUT.tsv INPUT.vcf.annotated Allele Gene Feature Feature_type Consequence cDNA_position CDS_position Protein_position Amino_acids Codons Existing_variation

Example input files can be found on synapse: ID syn11681983

Input file : Immunopeptidomes

  • A bed file containing the transcript name (ensembl transcript identifier), the start and end of the peptide that binds the MHC complex

Example immunopeptidome files can be found under the immunopeptidome directory

Output file

The output of SOPRANO consists on:

coverage ON_dnds ON_lowci ON_highci ON_muts OFF_dnds OFF_lowci OFF_highci OFF_muts P-val ON_na ON_NA ON_ns ON_NS OFF_na OFF_NA OFF_ns OFF_NS

coverage = Two options: ExonicOnly and ExonicIntronic. The latest should be used if there are intronic mutations in the mutation file. The algorithm uses intronic mutations to improve the background counts of silent mutations.

ON_dnds =  dN/dS of the target region provided in the bed file

ON_lowci = lower value for the 95% CI of the target

ON_highci = upper value for the 95% CI of the target

ON_muts = number of mutations observed inside the target region

OFF_dnds = dN/dS of the OFF-target region provided in the bed file

OFF_lowci = lower value for the 95% CI of the OFF-target

OFF_highci = upper value for the 95% CI of the OFF-target

OFF_muts = number of mutations observed outside the target region

P-val = P-value estimated from the comparison of the confidence intervals from ON and OFF dN/dS values

ON_na = Observed number of nonsilent mutations ON target

ON_NA = Number of nonsilent sites (corrected) ON target

ON_ns = Observed number of silent mutations ON target

ON_NS = Number of silent sites (corrected) ON target

OFF_na = Observed number of nonsilent mutations OFF target

OFF_NA = Number of nonsilent sites (corrected) OFF target

OFF_ns = Number of silent sites (corrected) OFF target

OFF_NS = Number of silent sites (corrected) OFF target

Notes before running

  • No header needed for input VEP file
  • VEP annotated first column must be in the format (chr_pos_ref/alt as described in https://www.ensembl.org/info/docs/tools/vep/vep_formats.html)
  • VEP annotated file must only have chromosomes that are 1,2,3,4...22 or uppercase X,Y
  • After you run vep with the option for ExAC frequencies, it would be necessary to remove all variants present in more than 0.1 percent of the population. You could apply the filter using: filter_vep -i input.annotated -f "ExAC_AF < 0.1 or not ExAC_AF" --ontology --filter "Consequence is coding_sequence_variant" -Be sure that you are using the GNU command line if you are running in a MacOS
  • Add dependencies to your path for easy running or hardcode the scripts
  • The genome file used in SOPRANO or SSB is a two column file that contains the info of the name of the fasta id (column 1) and the length of that sequence (column 2).
  • The UNIX system used should be able to recognize \t as a tab separator, some encodings may have problems on recognizing special characters
  • Earlier versions of bedtools will not work
  • Tab encoding should be \t (might be a problem for windows/OSX versions)
  • Genome length file is a two column file specifying the fasta id and the length of the sequence (see how to obtain it at the bottom)
  • Restrict your input dataset to chromosomes 1-22 and X and Y. Remove the rest.
  • Input chromosome number must coincide with reference genome and annotation (chr1 vs 1)

Genomes (Updated)

To get hg19 (GRCh37) fasta genome, you can download it from ensembl. Now we provide the matching chrom sizes file in the data directory:

wget https://ftp.ensembl.org/pub/grch37/current/fasta/homo_sapiens/dna/Homo_sapiens.GRCh37.dna.toplevel.fa.gz

GENOME=~/tools/SOPRANO/data/chrom_GRCh37.sizes

Obtain patient specific immune dN/dS values

To determine the patient specific immunopeptidome you should run the script get_epitope_HLA.pl:

As an example for :

perl scripts/get_epitope_HLA.pl examples/TCGA_hlaTypesTEST.tsv

This will create a command that you can run in a HPC cluster. In our example case, the command is:

egrep -w -e "HLA-A2402|HLA-A0301|HLA-B1501|HLA-B1801|HLA-C0701|HLA-C0303|" data/allhlaBinders_exprmean1.IEDBpeps.mgd.bed | sortBed -i stdin | mergeBed -i stdin > TCGA-HQ-A5ND.exprmean1.IEDBpeps.SB.epitope.bed
egrep -w -e "HLA-A0205|HLA-A3303|HLA-B5301|HLA-B5301|HLA-C0401|HLA-C0401|" data/allhlaBinders_exprmean1.IEDBpeps.mgd.bed | sortBed -i stdin | mergeBed -i stdin > TCGA-FD-A6TC.exprmean1.IEDBpeps.SB.epitope.bed

In this previous case, the global immunopeptidome (containing all possible HLA alleles) was filtered by expression and similarity to positive T-cell assays epitopes from IEDB as described in the manuscript. Alternatively, users can obtain an unfiltered global immunopeptidome (now provided in https://www.synapse.org/#!Synapse:syn30371735) and perform the post-processing of each patient specific immunopeptidome with user-defined filters.

After obtaining the immunopeptidome file following the steps before, you can run SOPRANO using the command following:

mkdir results_SOPRANO

./run_localSSBselection_v4.sh -i examples/TCGA-05-4396-01A-21D-1855-08.annotated -b immunopeptidomes/human/TCGA-05-4396.Expressed.IEDBpeps.SB.epitope.bedd -n TCGA-05-4396 -o results_SOPRANO -m ssb192

Limitations

  • SOPRANO will fail if there are 0 synonymous mutations inside the immunopeptidome.
  • Results must be taken cautiously when low number of synonymous mutations inside the immunopeptidome. See tutorial where we show the dependency of synonymous events versus the confidence interval length. In brief, one synonymous mutation in the immunopeptidome leads to a confidence interval length of ~0.6 and an 85% of misclassified cases
  • Only mutations in those transcript IDs present on the immunopeptidome file will be used to estimate ON and OFF dN/dS. The rest of mutations will be discarded.
  • SOPRANO can use mutations in intronic regions when using whole genome sequencing data to improve the estimate of the neutral mutation rate of the gene. However, this strategy is experimental since not every study will have information on all possible mutations falling into intronic regions (the length of the intronic region was estimated using hg19 reference).