SigProfilerAssignment

SigProfilerAssignment enables assignment of previously known mutational signatures to individual samples and individual somatic mutations. The tool refits different types of reference mutational signatures, including COSMIC signatures, as well as custom signature databases. Refitting of known mutational signatures is a numerical optimization approach that not only identifies the set of operative mutational signatures in a particular sample, but also quantifies the number of mutations assigned to each signature found in that sample. SigProfilerAssignment makes use of SigProfilerMatrixGenerator and SigProfilerPlotting, seamlessly integrating with other SigProfiler tools.

For users that prefer working in an R environment, a wrapper package is provided and can be found and installed from: https://github.com/AlexandrovLab/SigProfilerAssignmentR. Detailed documentation can be found at: https://osf.io/mz79v/wiki/home/.

Table of contents

• Installation
• Running
  • Main Parameters
  • Signature Subgroups
• Examples
• De novo extraction of mutational signatures downstream analysis
• Citation
• Copyright
• Contact Information

Installation

Install the current stable PyPi version of SigProfilerAssignment:

$ pip install SigProfilerAssignment

If mutation calling files (MAF, VCF, or simple text files) are used as input, please install your desired reference genome as follows (available reference genomes are: GRCh37, GRCh38, mm9, mm10, and rn6):

$ python
from SigProfilerMatrixGenerator import install as genInstall
genInstall.install('GRCh37')

Running

Assignment of known mutational signatures to individual samples is performed using the cosmic_fit function. Input samples are provided using the samples parameter in the form of mutation calling files (VCFs, MAFs, or simple text files), segmentation files or mutational matrices. COSMIC mutational signatures v3.4 are used as the default reference signatures, although previous COSMIC versions and custom signature databases are also supported using the cosmic_version and signature_database parameters. Results will be found in the folder specified in the output parameter.

from SigProfilerAssignment import Analyzer as Analyze
Analyze.cosmic_fit(samples, output, input_type="matrix", context_type="96",
                   collapse_to_SBS96=True, cosmic_version=3.4, exome=False,
                   genome_build="GRCh37", signature_database=None,
                   exclude_signature_subgroups=None, export_probabilities=False,
                   export_probabilities_per_mutation=False, make_plots=False,
                   sample_reconstruction_plots=False, verbose=False)

Main Parameters

Parameter	Variable Type	Parameter Description
samples	String	Path to the input somatic mutations file (if using segmentation file/mutational matrix) or input folder (mutation calling file/s).
output	String	Path to the output folder.
input_type	String	Three accepted input types: "vcf": if using mutation calling file/s (VCF, MAF, simple text file) as input "seg:TYPE": if using a segmentation file as input. Please check the required format at https://github.com/AlexandrovLab/SigProfilerMatrixGenerator#copy-number-matrix-generation. The accepted callers for TYPE are the following {"ASCAT", "ASCAT_NGS", "SEQUENZA", "ABSOLUTE", "BATTENBERG", "FACETS", "PURPLE", "TCGA"}. For example:"seg:BATTENBERG" "matrix": if using a mutational matrix as input The default value is "matrix".
context_type	String	Required context type if input_type is "vcf". context_type takes which context type of the input data is considered for assignment. Valid options include "96", "288", "1536", "DINUC", and "ID". The default value is "96".
cosmic_version	Float	Defines the version of the COSMIC reference signatures. Takes a positive float among 1, 2, 3, 3.1, 3.2, 3.3, and 3.4. The default value is 3.4.
exome	Boolean	Defines if the exome renormalized COSMIC signatures will be used. The default value is False.
genome_build	String	The reference genome build, used for select the appropriate version of the COSMIC reference signatures, as well as processing the mutation calling file/s. Supported genomes include "GRCh37", "GRCh38", "mm9", "mm10" and "rn6". The default value is "GRCh37". If the selected genome is not in the supported list, the default genome will be used.
signature_database	String	Path to the input set of known mutational signatures (only in case that COSMIC reference signatures are not used), a tab delimited file that contains the signature matrix where the rows are mutation types and columns are signature IDs.
exclude_signature_subgroups	List	Removes the signatures corresponding to specific subtypes to improve refitting (only available when using default COSMIC reference signatures). The usage is explained below. The default value is None, which corresponds to use all COSMIC signatures.
export_probabilities	Boolean	Defines if the probability matrix per mutational context for all samples is created. The default value is True.
export_probabilities_per_mutation	Boolean	Defines if the probability matrices per mutation for all samples are created. Only available when input_type is "vcf". The default value is False.
make_plots	Boolean	Toggle on and off for making and saving plots. The default value is True.
sample_reconstruction_plots	String	Select the output format for sample reconstruction plots. Valid inputs are {'pdf', 'png', 'both', None}. The default value is None.
verbose	Boolean	Prints detailed statements. The default value is False.
volume	String	Path to SigProfilerAssignment volumes. Used for Docker/Singularity. Environmental variable "SIGPROFILERASSIGNMENT_VOLUME" takes precedence. Default value is None.

Signature Subgroups

When using COSMIC reference signatures, some subgroups of signatures can be removed to improve the refitting analysis. To use this feature, the exclude_signature_subgroups parameter should be added, following the sintax below:

exclude_signature_subgroups = ['MMR_deficiency_signatures',
                               'POL_deficiency_signatures',
                               'HR_deficiency_signatures' ,
                               'BER_deficiency_signatures',
                               'Chemotherapy_signatures',
                               'Immunosuppressants_signatures'
                               'Treatment_signatures'
                               'APOBEC_signatures',
                               'Tobacco_signatures',
                               'UV_signatures',
                               'AA_signatures',
                               'Colibactin_signatures',
                               'Artifact_signatures',
                               'Lymphoid_signatures']

The full list of signature subgroups is included in the following table:

Signature subgroup	SBS signatures excluded	DBS signatures excluded	ID signatures excluded
MMR_deficiency_signatures	6, 14, 15, 20, 21, 26, 44	7, 10	7
POL_deficiency_signatures	10a, 10b, 10c, 10d, 28	3	-
HR_deficiency_signatures	3	13	6
BER_deficiency_signatures	30, 36	-	-
Chemotherapy_signatures	11, 25, 31, 35, 86, 87, 90, 99	5	-
Immunosuppressants_signatures	32	-	-
Treatment_signatures	11, 25, 31, 32, 35, 86, 87, 90, 99	5	-
APOBEC_signatures	2, 13	-	-
Tobacco_signatures	4, 29, 92	2	3
UV_signatures	7a, 7b, 7c, 7d, 38	1	13
AA_signatures	22a, 22b	20	23
Colibactin_signatures	88	-	18
Artifact_signatures	27, 43, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 95	14	-
Lymphoid_signatures	9, 84, 85	-	-

Examples

Using mutation calling files (VCFs) as input

import SigProfilerAssignment as spa
from SigProfilerAssignment import Analyzer as Analyze

Analyze.cosmic_fit(samples=spa.__path__[0]+"/data/tests/vcf_input", 
                   output="example_vcf",
                   input_type="vcf",
                   context_type="96",
                   genome_build="GRCh37",
                   cosmic_version=3.4)

Using a multi-sample segmentation file as input

import SigProfilerAssignment as spa
from SigProfilerAssignment import Analyzer as Analyze

Analyze.cosmic_fit(samples=spa.__path__[0]+"/data/tests/cnv_input/all.breast.ascat.summary.sample.tsv", 
                   output="example_sf",
                   input_type="seg:ASCAT_NGS",
                   cosmic_version=3.4,
                   collapse_to_SBS96=False)

Using a mutational matrix as input

import SigProfilerAssignment as spa
from SigProfilerAssignment import Analyzer as Analyze

Analyze.cosmic_fit(samples=spa.__path__[0]+"/data/tests/txt_input/sample_matrix_SBS.txt", 
                   output="example_mm",
                   input_type="matrix",
                   genome_build="GRCh37",
                   cosmic_version=3.4)

De novo extraction of mutational signatures downstream analysis

Additional functionalities for downstream analysis of de novo extraction of mutational signatures are also available as part of SigProfilerAssignment, including assignment of de novo extracted mutational signatures and decomposition of de novo signatures using a known set of signatures. More information can be found on the wiki page at https://osf.io/mz79v/wiki/5.%20Advanced%20mode/.

Unit Tests

Unit tests can be run with the following commands:

python setup.py sdist
pip install .[tests]
pytest tests

Citation

Díaz-Gay, M., Vangara, R., Barnes, M., ... & Alexandrov, L. B. (2023). Assigning mutational signatures to individual samples and individual somatic mutations with SigProfilerAssignment, Bioinformatics, 2023-07. doi: https://doi.org/10.1093/bioinformatics/btad756

Copyright

This software and its documentation are copyright 2022 as a part of the SigProfiler project. The SigProfilerAssignment framework is free software and is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

Contact Information

Please address any queries or bug reports to Raviteja Vangara at [email protected] or Marcos Díaz-Gay at [email protected].