Abstract B012: Predicting targetable paralog synthetic lethalities and functional redundancies in cancer genomes
Bibliographic record
Abstract
Abstract Paralogs are prevalent in the human genome and are considered a rich source of synthetic lethality due to functional redundancy. For a cancer cell carrying a gene with Loss-Of-Function (LOF) mutation, inactivation of its paralog using gene editing can induce a selective decrease in viability, leaving normal cells that do not harbor the mutation unharmed. Previous studies have exploited this vulnerability of cancer genomes. However, the cancer-specificity of such synthetic lethality limits its application across different cancer types. Furthermore, the role of functional redundancy which can explain the cancer specificity has remained largely unexplored. In this study, we computationally predicted synthetic lethal paralogs along with mechanistically important functional redundancies between them in a cancer-specific manner. We applied our prediction method to publicly available data for an aggressive subtype of breast cancer called triple-negative breast cancer (TNBC), which lacks the expression key biomarkers and hence has the worst prognosis among other breast cancer subtypes. TNBC is characterized by the highest mutational load and the largest fraction of genome altered among the breast cancer subtypes providing a rich mutational landscape to identify LOF genes. Using the CRISPR inactivation screen data for cancer cell lines obtained from the Cancer Dependency Map (DepMap) project, we predicted sets of synthetic lethal paralogs that show a significantly greater viability decrease if a gene carries LOF and its paralog is inactivated, compared to the viability decrease due to the inactivation of only one of the paralogs. Consistent with previous findings of context-dependent synthetic lethality, we found that a relatively small fraction of TNBC-specific synthetic lethal paralogs overlapped with those found across other cancer types. To uncover the mechanistically important functional redundancies between paralogs, we analyzed the genomics and transcriptomics data from multiple sources: TNBC panel of primary tumors and patient-derived xenografts, Pan-Cancer Analysis of Whole Genomes (PCAWG), and the Cancer Cell Line Encyclopedia (CCLE). The functional redundancies varied across cancer types based on (1) mutual exclusivity of LOFs, (2) backup compensation of deleterious mutations, (3) backup upregulation, and (4) dosage balance. Overall, our findings show a strong context-dependency of synthetic lethal paralogs and estimates of functional redundancy, emphasizing the importance of such cancer-specific predictions in identifying targetable paralog synthetic lethalities. Collectively, the computational method and sets of targetable paralogs are a unique resource for developing precision oncology therapeutic strategies against TNBC and other cancers more broadly. Citation Format: Rohan Dandage, Elena Kuzmin. Predicting targetable paralog synthetic lethalities and functional redundancies in cancer genomes [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Expanding and Translating Cancer Synthetic Vulnerabilities; 2024 Jun 10-13; Montreal, Quebec, Canada. Philadelphia (PA): AACR; Mol Cancer Ther 2024;23(6 Suppl):Abstract nr B012.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".