1
|
Pancotti C, Rollo C, Codicè F, Birolo G, Fariselli P, Sanavia T. MUSE-XAE: MUtational Signature Extraction with eXplainable AutoEncoder enhances tumour types classification. Bioinformatics 2024; 40:btae320. [PMID: 38754097 PMCID: PMC11139523 DOI: 10.1093/bioinformatics/btae320] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/28/2023] [Revised: 04/08/2024] [Accepted: 05/15/2024] [Indexed: 05/18/2024] Open
Abstract
MOTIVATION Mutational signatures are a critical component in deciphering the genetic alterations that underlie cancer development and have become a valuable resource to understand the genomic changes during tumorigenesis. Therefore, it is essential to employ precise and accurate methods for their extraction to ensure that the underlying patterns are reliably identified and can be effectively utilized in new strategies for diagnosis, prognosis, and treatment of cancer patients. RESULTS We present MUSE-XAE, a novel method for mutational signature extraction from cancer genomes using an explainable autoencoder. Our approach employs a hybrid architecture consisting of a nonlinear encoder that can capture nonlinear interactions among features, and a linear decoder which ensures the interpretability of the active signatures. We evaluated and compared MUSE-XAE with other available tools on both synthetic and real cancer datasets and demonstrated that it achieves superior performance in terms of precision and sensitivity in recovering mutational signature profiles. MUSE-XAE extracts highly discriminative mutational signature profiles by enhancing the classification of primary tumour types and subtypes in real world settings. This approach could facilitate further research in this area, with neural networks playing a critical role in advancing our understanding of cancer genomics. AVAILABILITY AND IMPLEMENTATION MUSE-XAE software is freely available at https://github.com/compbiomed-unito/MUSE-XAE.
Collapse
Affiliation(s)
- Corrado Pancotti
- Computational Biomedicine Unit, Department of Medical Sciences, University of Torino, via Santena 19, Torino 10126, Italy
| | - Cesare Rollo
- Computational Biomedicine Unit, Department of Medical Sciences, University of Torino, via Santena 19, Torino 10126, Italy
| | - Francesco Codicè
- Computational Biomedicine Unit, Department of Medical Sciences, University of Torino, via Santena 19, Torino 10126, Italy
| | - Giovanni Birolo
- Computational Biomedicine Unit, Department of Medical Sciences, University of Torino, via Santena 19, Torino 10126, Italy
| | - Piero Fariselli
- Computational Biomedicine Unit, Department of Medical Sciences, University of Torino, via Santena 19, Torino 10126, Italy
| | - Tiziana Sanavia
- Computational Biomedicine Unit, Department of Medical Sciences, University of Torino, via Santena 19, Torino 10126, Italy
| |
Collapse
|
2
|
Jin H, Gulhan DC, Geiger B, Ben-Isvy D, Geng D, Ljungström V, Park PJ. Accurate and sensitive mutational signature analysis with MuSiCal. Nat Genet 2024; 56:541-552. [PMID: 38361034 PMCID: PMC10937379 DOI: 10.1038/s41588-024-01659-0] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/16/2022] [Accepted: 01/08/2024] [Indexed: 02/17/2024]
Abstract
Mutational signature analysis is a recent computational approach for interpreting somatic mutations in the genome. Its application to cancer data has enhanced our understanding of mutational forces driving tumorigenesis and demonstrated its potential to inform prognosis and treatment decisions. However, methodological challenges remain for discovering new signatures and assigning proper weights to existing signatures, thereby hindering broader clinical applications. Here we present Mutational Signature Calculator (MuSiCal), a rigorous analytical framework with algorithms that solve major problems in the standard workflow. Our simulation studies demonstrate that MuSiCal outperforms state-of-the-art algorithms for both signature discovery and assignment. By reanalyzing more than 2,700 cancer genomes, we provide an improved catalog of signatures and their assignments, discover nine indel signatures absent in the current catalog, resolve long-standing issues with the ambiguous 'flat' signatures and give insights into signatures with unknown etiologies. We expect MuSiCal and the improved catalog to be a step towards establishing best practices for mutational signature analysis.
Collapse
Affiliation(s)
- Hu Jin
- Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA
| | - Doga C Gulhan
- Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA
| | - Benedikt Geiger
- Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA
| | - Daniel Ben-Isvy
- Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA
| | - David Geng
- Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA
| | - Viktor Ljungström
- Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA
| | - Peter J Park
- Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA.
| |
Collapse
|
3
|
Ostroverkhova D, Przytycka TM, Panchenko AR. Cancer driver mutations: predictions and reality. Trends Mol Med 2023:S1471-4914(23)00067-9. [PMID: 37076339 DOI: 10.1016/j.molmed.2023.03.007] [Citation(s) in RCA: 10] [Impact Index Per Article: 10.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/03/2023] [Revised: 03/17/2023] [Accepted: 03/23/2023] [Indexed: 04/21/2023]
Abstract
Cancer cells accumulate many genetic alterations throughout their lifetime, but only a few of them drive cancer progression, termed driver mutations. Driver mutations may vary between cancer types and patients, can remain latent for a long time and become drivers at particular cancer stages, or may drive oncogenesis only in conjunction with other mutations. The high mutational, biochemical, and histological tumor heterogeneity makes driver mutation identification very challenging. In this review we summarize recent efforts to identify driver mutations in cancer and annotate their effects. We underline the success of computational methods to predict driver mutations in finding novel cancer biomarkers, including in circulating tumor DNA (ctDNA). We also report on the boundaries of their applicability in clinical research.
Collapse
Affiliation(s)
- Daria Ostroverkhova
- Department of Pathology and Molecular Medicine, Queen's University, Kingston, ON, Canada
| | - Teresa M Przytycka
- National Library of Medicine, National Institutes of Health (NIH), Bethesda, MD, USA.
| | - Anna R Panchenko
- Department of Pathology and Molecular Medicine, Queen's University, Kingston, ON, Canada; Department of Biology and Molecular Sciences, Queen's University, Kingston, ON, Canada; School of Computing, Queen's University, Kingston, ON, Canada; Ontario Institute of Cancer Research, Toronto, ON, Canada.
| |
Collapse
|
4
|
Mas-Ponte D, McCullough M, Supek F. Spectrum of DNA mismatch repair failures viewed through the lens of cancer genomics and implications for therapy. Clin Sci (Lond) 2022; 136:383-404. [PMID: 35274136 PMCID: PMC8919091 DOI: 10.1042/cs20210682] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/01/2021] [Revised: 02/02/2022] [Accepted: 02/28/2022] [Indexed: 12/15/2022]
Abstract
Genome sequencing can be used to detect DNA repair failures in tumors and learn about underlying mechanisms. Here, we synthesize findings from genomic studies that examined deficiencies of the DNA mismatch repair (MMR) pathway. The impairment of MMR results in genome-wide hypermutation and in the 'microsatellite instability' (MSI) phenotype-occurrence of indel mutations at short tandem repeat (microsatellite) loci. The MSI status of tumors was traditionally assessed by molecular testing of a selected set of MS loci or by measuring MMR protein expression levels. Today, genomic data can provide a more complete picture of the consequences on genomic instability. Multiple computational studies examined somatic mutation distributions that result from failed DNA repair pathways in tumors. These include analyzing the commonly studied trinucleotide mutational spectra of single-nucleotide variants (SNVs), as well as of other features such as indels, structural variants, mutation clusters and regional mutation rate redistribution. The identified mutation patterns can be used to rigorously measure prevalence of MMR failures across cancer types, and potentially to subcategorize the MMR deficiencies. Diverse data sources, genomic and pre-genomic, from human and from experimental models, suggest there are different ways in which MMR can fail, and/or that the cell-type or genetic background may result in different types of MMR mutational patterns. The spectrum of MMR failures may direct cancer evolution, generating particular sets of driver mutations. Moreover, MMR affects outcomes of therapy by DNA damaging drugs, antimetabolites, nonsense-mediated mRNA decay (NMD) inhibitors, and immunotherapy by promoting either resistance or sensitivity, depending on the type of therapy.
Collapse
Affiliation(s)
- David Mas-Ponte
- Genome Data Science, Institute for Research in Biomedicine (IRB Barcelona), The Barcelona Institute for Science and Technology, Baldiri Reixac 10, Barcelona 08028, Spain
| | - Marcel McCullough
- Genome Data Science, Institute for Research in Biomedicine (IRB Barcelona), The Barcelona Institute for Science and Technology, Baldiri Reixac 10, Barcelona 08028, Spain
| | - Fran Supek
- Genome Data Science, Institute for Research in Biomedicine (IRB Barcelona), The Barcelona Institute for Science and Technology, Baldiri Reixac 10, Barcelona 08028, Spain
- Catalan Institution for Research and Advanced Studies (ICREA), Pg Lluís Companys, 23, Barcelona 08010, Spain
| |
Collapse
|
5
|
Bernardo S, Meyenberg M, Loizou JI. Decomposing the mutational landscape of cancer genomes with RepairSig. Cell Syst 2021; 12:953-955. [PMID: 34672958 DOI: 10.1016/j.cels.2021.09.009] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/20/2022]
Abstract
Mutational signatures are the outcomes of mutagenic processes that occur prior to, and during, tumorigenesis as a result of DNA damage, DNA repair, and DNA replication. In this issue of Cell Systems, Wojtowicz et al. introduce a new computational model aimed at deconstructing the mutational processes that shape cancer genomes.
Collapse
Affiliation(s)
- Sara Bernardo
- Institute of Cancer Research, Department of Medicine I, Comprehensive Cancer Centre, Medical University of Vienna, Borschkegasse 8a, 1090 Vienna, Austria.
| | - Mathilde Meyenberg
- Institute of Cancer Research, Department of Medicine I, Comprehensive Cancer Centre, Medical University of Vienna, Borschkegasse 8a, 1090 Vienna, Austria; CeMM Research Center for Molecular Medicine of the Austrian Academy of Sciences, Vienna, Austria
| | - Joanna I Loizou
- Institute of Cancer Research, Department of Medicine I, Comprehensive Cancer Centre, Medical University of Vienna, Borschkegasse 8a, 1090 Vienna, Austria; CeMM Research Center for Molecular Medicine of the Austrian Academy of Sciences, Vienna, Austria.
| |
Collapse
|