CS
C.F. Seale
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
7 records found
1
Many tumours show deficiencies in DNA damage response (DDR), not only driving tumorigenesis but also exposing vulnerabilities with therapeutic potential. Assessing which patients might benefit from DDR-targeting therapy requires knowledge of tumour DDR deficiency (DDRd) status, with mutational signatures reportedly better predictors than loss-of-function mutations. Existing DDRd models offer effective prediction for pathways with well-characterized processes and mutational signatures. Nevertheless, development of models for additional DDRd and clinically relevant mechanisms could be hampered by the fact that most mutational signatures have unknown etiology. Using supervised non-negative matrix factorization (SNMF), we integrate mutational signature learning with multiclass DDR-deficiency prediction to enable etiology-guided learning of signatures from cell lines with confirmed gene knockouts. Applied to DDR gene knockout human-induced pluripotent stem cell lines, SNMF identified etiology-aligned representations of deficiency in homologous recombination, mismatch repair, and base excision repair. Even guided by pathway-level labels, SNMF captured gene-specific base excision repair submechanisms, showing the integration offered added granularity. Learned cell line signatures showed high similarity to tumour-derived COSMIC signatures, revealed associations with mutations in DDR genes, and enabled high recall of tumours with DDR deficiencies. We envision that SNMF-like methods could leverage knockout screens to learn etiology-guided signatures for improved DDRd annotation and treatment optimization. SNMF is available at: https://github.com/joanagoncalveslab/SNMF.
...
Many tumours show deficiencies in DNA damage response (DDR), not only driving tumorigenesis but also exposing vulnerabilities with therapeutic potential. Assessing which patients might benefit from DDR-targeting therapy requires knowledge of tumour DDR deficiency (DDRd) status, with mutational signatures reportedly better predictors than loss-of-function mutations. Existing DDRd models offer effective prediction for pathways with well-characterized processes and mutational signatures. Nevertheless, development of models for additional DDRd and clinically relevant mechanisms could be hampered by the fact that most mutational signatures have unknown etiology. Using supervised non-negative matrix factorization (SNMF), we integrate mutational signature learning with multiclass DDR-deficiency prediction to enable etiology-guided learning of signatures from cell lines with confirmed gene knockouts. Applied to DDR gene knockout human-induced pluripotent stem cell lines, SNMF identified etiology-aligned representations of deficiency in homologous recombination, mismatch repair, and base excision repair. Even guided by pathway-level labels, SNMF captured gene-specific base excision repair submechanisms, showing the integration offered added granularity. Learned cell line signatures showed high similarity to tumour-derived COSMIC signatures, revealed associations with mutations in DDR genes, and enabled high recall of tumours with DDR deficiencies. We envision that SNMF-like methods could leverage knockout screens to learn etiology-guided signatures for improved DDRd annotation and treatment optimization. SNMF is available at: https://github.com/joanagoncalveslab/SNMF.
Journal article
(2026)
-
C.F. Seale, Marco Barazas, Robin van Schendel, Marcel Tijsterman, Joana P. Gonçalves
Understanding DNA double-strand break (DSB) repair is crucial for the development of targeted anticancer therapies, yet the roles of many genes remain unclear. Recent studies show that disruption of known DSB repair genes can alter the sequence-specific distribution of mutations arising after DSB repair, suggesting that genome-wide perturbation screens could be leveraged to identify new DSB genes leading to distinct deviations from the expected wild-type distribution. Given the challenges of designing controls for a genome-wide screen, we explore the high gene throughput to forgo the use of traditional controls by reframing the analysis as an outlier detection problem, assuming that most genes have minimal influence on DSB repair outcomes. We propose MUSICiAn (Mutational Signature Catalogue Analysis), a compositional data analysis method that ranks gene perturbation impact on mutational spectra without controls by measuring deviations from the central tendency considering the distribution of all spectra. We show that MUSICiAn effectively estimates pseudo-controls for the Repair-seq screen, yielding 476 genes and 60 nontargeting controls. We further apply MUSICiAn to the first genome-wide screen of 18 406 genes with mutational spectra readout, MUSIC, reporting that MUSICiAn successfully recovers known DSB repair genes, highlights the spliceosome as a lesser-appreciated player, and reveals candidates for further investigation.
...
Understanding DNA double-strand break (DSB) repair is crucial for the development of targeted anticancer therapies, yet the roles of many genes remain unclear. Recent studies show that disruption of known DSB repair genes can alter the sequence-specific distribution of mutations arising after DSB repair, suggesting that genome-wide perturbation screens could be leveraged to identify new DSB genes leading to distinct deviations from the expected wild-type distribution. Given the challenges of designing controls for a genome-wide screen, we explore the high gene throughput to forgo the use of traditional controls by reframing the analysis as an outlier detection problem, assuming that most genes have minimal influence on DSB repair outcomes. We propose MUSICiAn (Mutational Signature Catalogue Analysis), a compositional data analysis method that ranks gene perturbation impact on mutational spectra without controls by measuring deviations from the central tendency considering the distribution of all spectra. We show that MUSICiAn effectively estimates pseudo-controls for the Repair-seq screen, yielding 476 genes and 60 nontargeting controls. We further apply MUSICiAn to the first genome-wide screen of 18 406 genes with mutational spectra readout, MUSIC, reporting that MUSICiAn successfully recovers known DSB repair genes, highlights the spliceosome as a lesser-appreciated player, and reveals candidates for further investigation.
CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) technology has transformed molecular biology by enabling a strategy for precise, efficient, and relatively simple genome editing. Guided by a small strand of RNA, CRISPR locates specific DNA sequences within the genome and introduces double-strand breaks (DSBs). A typical cell can detect and fix the damage by invoking one of several DNA repair pathways. However, repair is not error-free and often introduces mutations. The mutagenic nature of repair pathways can be leveraged to disrupt genes or regulatory elements with high specificity, providing a powerful tool for gaining insights into gene function. Researchers can also generate datasets of mutations left behind after DSB induction and repair within different genomic contexts to learn more about the mutagenic effects of DNA repair. In this thesis, we explore challenges and novel approaches for analysing large-scale datasets of mutations and gene essentiality generated via CRISPR technology.
...
CRISPR (Clustered Regularly Interspaced Short Palindromic Repeats) technology has transformed molecular biology by enabling a strategy for precise, efficient, and relatively simple genome editing. Guided by a small strand of RNA, CRISPR locates specific DNA sequences within the genome and introduces double-strand breaks (DSBs). A typical cell can detect and fix the damage by invoking one of several DNA repair pathways. However, repair is not error-free and often introduces mutations. The mutagenic nature of repair pathways can be leveraged to disrupt genes or regulatory elements with high specificity, providing a powerful tool for gaining insights into gene function. Researchers can also generate datasets of mutations left behind after DSB induction and repair within different genomic contexts to learn more about the mutagenic effects of DNA repair. In this thesis, we explore challenges and novel approaches for analysing large-scale datasets of mutations and gene essentiality generated via CRISPR technology.
Motivation Controlling the outcomes of CRISPR editing is crucial for the success of gene therapy. Since donor template-based editing is often inefficient, alternative strategies have emerged that leverage mutagenic end-joining repair instead. Existing machine learning models can accurately predict end-joining repair outcomes; however, generalisability beyond the specific cell line used for training remains a challenge, and interpretability is typically limited by suboptimal feature representation and model architecture. Results We propose X-CRISP, a flexible and interpretable neural network for predicting repair outcome frequencies based on a minimal set of outcome and sequence features, including microhomologies (MH). Outperforming prior models on detailed and aggregate outcome predictions, X-CRISP prioritised MH location over MH sequence properties such as GC content for deletion outcomes. Through transfer learning, we adapted X-CRISP pre-trained on wild-type mESC data to target human cell lines K562, HAP1, U2OS, and mESC lines with altered DNA repair function. Adapted X-CRISP models improved over direct training on target data from as few as 50 samples, suggesting that this strategy could be leveraged to build models for new domains using a fraction of the data required to train models from scratch.
...
Motivation Controlling the outcomes of CRISPR editing is crucial for the success of gene therapy. Since donor template-based editing is often inefficient, alternative strategies have emerged that leverage mutagenic end-joining repair instead. Existing machine learning models can accurately predict end-joining repair outcomes; however, generalisability beyond the specific cell line used for training remains a challenge, and interpretability is typically limited by suboptimal feature representation and model architecture. Results We propose X-CRISP, a flexible and interpretable neural network for predicting repair outcome frequencies based on a minimal set of outcome and sequence features, including microhomologies (MH). Outperforming prior models on detailed and aggregate outcome predictions, X-CRISP prioritised MH location over MH sequence properties such as GC content for deletion outcomes. Through transfer learning, we adapted X-CRISP pre-trained on wild-type mESC data to target human cell lines K562, HAP1, U2OS, and mESC lines with altered DNA repair function. Adapted X-CRISP models improved over direct training on target data from as few as 50 samples, suggesting that this strategy could be leveraged to build models for new domains using a fraction of the data required to train models from scratch.
Motivation
Anti-cancer therapies based on synthetic lethality (SL) exploit tumour vulnerabilities for treatment with reduced side effects, by targeting a gene that is jointly essential with another whose function is lost. Computational prediction is key to expedite SL screening, yet existing methods are vulnerable to prevalent selection bias in SL data and reliant on cancer or tissue type-specific omics, which can be scarce. Notably, sequence similarity remains underexplored as a proxy for related gene function and joint essentiality.
Results
We propose ELISL, Early–Late Integrated SL prediction with forest ensembles, using context-free protein sequence embeddings and context-specific omics from cell lines and tissue. Across eight cancer types, ELISL showed superior robustness to selection bias and recovery of known SL genes, as well as promising cross-cancer predictions. Co-occurring mutations in a BRCA gene and ELISL-predicted pairs from the HH, FGF, WNT, or NEIL gene families were associated with longer patient survival times, revealing therapeutic potential. ...
Anti-cancer therapies based on synthetic lethality (SL) exploit tumour vulnerabilities for treatment with reduced side effects, by targeting a gene that is jointly essential with another whose function is lost. Computational prediction is key to expedite SL screening, yet existing methods are vulnerable to prevalent selection bias in SL data and reliant on cancer or tissue type-specific omics, which can be scarce. Notably, sequence similarity remains underexplored as a proxy for related gene function and joint essentiality.
Results
We propose ELISL, Early–Late Integrated SL prediction with forest ensembles, using context-free protein sequence embeddings and context-specific omics from cell lines and tissue. Across eight cancer types, ELISL showed superior robustness to selection bias and recovery of known SL genes, as well as promising cross-cancer predictions. Co-occurring mutations in a BRCA gene and ELISL-predicted pairs from the HH, FGF, WNT, or NEIL gene families were associated with longer patient survival times, revealing therapeutic potential. ...
Motivation
Anti-cancer therapies based on synthetic lethality (SL) exploit tumour vulnerabilities for treatment with reduced side effects, by targeting a gene that is jointly essential with another whose function is lost. Computational prediction is key to expedite SL screening, yet existing methods are vulnerable to prevalent selection bias in SL data and reliant on cancer or tissue type-specific omics, which can be scarce. Notably, sequence similarity remains underexplored as a proxy for related gene function and joint essentiality.
Results
We propose ELISL, Early–Late Integrated SL prediction with forest ensembles, using context-free protein sequence embeddings and context-specific omics from cell lines and tissue. Across eight cancer types, ELISL showed superior robustness to selection bias and recovery of known SL genes, as well as promising cross-cancer predictions. Co-occurring mutations in a BRCA gene and ELISL-predicted pairs from the HH, FGF, WNT, or NEIL gene families were associated with longer patient survival times, revealing therapeutic potential.
Anti-cancer therapies based on synthetic lethality (SL) exploit tumour vulnerabilities for treatment with reduced side effects, by targeting a gene that is jointly essential with another whose function is lost. Computational prediction is key to expedite SL screening, yet existing methods are vulnerable to prevalent selection bias in SL data and reliant on cancer or tissue type-specific omics, which can be scarce. Notably, sequence similarity remains underexplored as a proxy for related gene function and joint essentiality.
Results
We propose ELISL, Early–Late Integrated SL prediction with forest ensembles, using context-free protein sequence embeddings and context-specific omics from cell lines and tissue. Across eight cancer types, ELISL showed superior robustness to selection bias and recovery of known SL genes, as well as promising cross-cancer predictions. Co-occurring mutations in a BRCA gene and ELISL-predicted pairs from the HH, FGF, WNT, or NEIL gene families were associated with longer patient survival times, revealing therapeutic potential.
Motivation
Synthetic lethality (SL) between two genes occurs when simultaneous loss of function leads to cell death. This holds great promise for developing anti-cancer therapeutics that target synthetic lethal pairs of endogenously disrupted genes. Identifying novel SL relationships through exhaustive experimental screens is challenging, due to the vast number of candidate pairs. Computational SL prediction is therefore sought to identify promising SL gene pairs for further experimentation. However, current SL prediction methods lack consideration for generalizability in the presence of selection bias in SL data.
Results
We show that SL data exhibit considerable gene selection bias. Our experiments designed to assess the robustness of SL prediction reveal that models driven by the topology of known SL interactions (e.g. graph, matrix factorization) are especially sensitive to selection bias. We introduce selection bias-resilient synthetic lethality (SBSL) prediction using regularized logistic regression or random forests. Each gene pair is described by 27 molecular features derived from cancer cell line, cancer patient tissue and healthy donor tissue samples. SBSL models are built and tested using approximately 8000 experimentally derived SL pairs across breast, colon, lung and ovarian cancers. Compared to other SL prediction methods, SBSL showed higher predictive performance, better generalizability and robustness to selection bias. Gene dependency, quantifying the essentiality of a gene for cell survival, contributed most to SBSL predictions. Random forests were superior to linear models in the absence of dependency features, highlighting the relevance of mutual exclusivity of somatic mutations, co-expression in healthy tissue and differential expression in tumour samples.
Availability and implementation
https://github.com/joanagoncalveslab/sbsl
Supplementary information
Supplementary data are available at Bioinformatics online. ...
Synthetic lethality (SL) between two genes occurs when simultaneous loss of function leads to cell death. This holds great promise for developing anti-cancer therapeutics that target synthetic lethal pairs of endogenously disrupted genes. Identifying novel SL relationships through exhaustive experimental screens is challenging, due to the vast number of candidate pairs. Computational SL prediction is therefore sought to identify promising SL gene pairs for further experimentation. However, current SL prediction methods lack consideration for generalizability in the presence of selection bias in SL data.
Results
We show that SL data exhibit considerable gene selection bias. Our experiments designed to assess the robustness of SL prediction reveal that models driven by the topology of known SL interactions (e.g. graph, matrix factorization) are especially sensitive to selection bias. We introduce selection bias-resilient synthetic lethality (SBSL) prediction using regularized logistic regression or random forests. Each gene pair is described by 27 molecular features derived from cancer cell line, cancer patient tissue and healthy donor tissue samples. SBSL models are built and tested using approximately 8000 experimentally derived SL pairs across breast, colon, lung and ovarian cancers. Compared to other SL prediction methods, SBSL showed higher predictive performance, better generalizability and robustness to selection bias. Gene dependency, quantifying the essentiality of a gene for cell survival, contributed most to SBSL predictions. Random forests were superior to linear models in the absence of dependency features, highlighting the relevance of mutual exclusivity of somatic mutations, co-expression in healthy tissue and differential expression in tumour samples.
Availability and implementation
https://github.com/joanagoncalveslab/sbsl
Supplementary information
Supplementary data are available at Bioinformatics online. ...
Motivation
Synthetic lethality (SL) between two genes occurs when simultaneous loss of function leads to cell death. This holds great promise for developing anti-cancer therapeutics that target synthetic lethal pairs of endogenously disrupted genes. Identifying novel SL relationships through exhaustive experimental screens is challenging, due to the vast number of candidate pairs. Computational SL prediction is therefore sought to identify promising SL gene pairs for further experimentation. However, current SL prediction methods lack consideration for generalizability in the presence of selection bias in SL data.
Results
We show that SL data exhibit considerable gene selection bias. Our experiments designed to assess the robustness of SL prediction reveal that models driven by the topology of known SL interactions (e.g. graph, matrix factorization) are especially sensitive to selection bias. We introduce selection bias-resilient synthetic lethality (SBSL) prediction using regularized logistic regression or random forests. Each gene pair is described by 27 molecular features derived from cancer cell line, cancer patient tissue and healthy donor tissue samples. SBSL models are built and tested using approximately 8000 experimentally derived SL pairs across breast, colon, lung and ovarian cancers. Compared to other SL prediction methods, SBSL showed higher predictive performance, better generalizability and robustness to selection bias. Gene dependency, quantifying the essentiality of a gene for cell survival, contributed most to SBSL predictions. Random forests were superior to linear models in the absence of dependency features, highlighting the relevance of mutual exclusivity of somatic mutations, co-expression in healthy tissue and differential expression in tumour samples.
Availability and implementation
https://github.com/joanagoncalveslab/sbsl
Supplementary information
Supplementary data are available at Bioinformatics online.
Synthetic lethality (SL) between two genes occurs when simultaneous loss of function leads to cell death. This holds great promise for developing anti-cancer therapeutics that target synthetic lethal pairs of endogenously disrupted genes. Identifying novel SL relationships through exhaustive experimental screens is challenging, due to the vast number of candidate pairs. Computational SL prediction is therefore sought to identify promising SL gene pairs for further experimentation. However, current SL prediction methods lack consideration for generalizability in the presence of selection bias in SL data.
Results
We show that SL data exhibit considerable gene selection bias. Our experiments designed to assess the robustness of SL prediction reveal that models driven by the topology of known SL interactions (e.g. graph, matrix factorization) are especially sensitive to selection bias. We introduce selection bias-resilient synthetic lethality (SBSL) prediction using regularized logistic regression or random forests. Each gene pair is described by 27 molecular features derived from cancer cell line, cancer patient tissue and healthy donor tissue samples. SBSL models are built and tested using approximately 8000 experimentally derived SL pairs across breast, colon, lung and ovarian cancers. Compared to other SL prediction methods, SBSL showed higher predictive performance, better generalizability and robustness to selection bias. Gene dependency, quantifying the essentiality of a gene for cell survival, contributed most to SBSL predictions. Random forests were superior to linear models in the absence of dependency features, highlighting the relevance of mutual exclusivity of somatic mutations, co-expression in healthy tissue and differential expression in tumour samples.
Availability and implementation
https://github.com/joanagoncalveslab/sbsl
Supplementary information
Supplementary data are available at Bioinformatics online.