Computational methods, databases and tools for synthetic lethality prediction.

Area of Science:

• Genetics and Genomics
• Computational Biology
• Cancer Therapeutics

Background:

• Synthetic lethality (SL) is a genetic interaction where simultaneous inactivation of two genes leads to cell death, while individual inactivation is tolerated.
• SL-based therapies, exemplified by PARP inhibitors, represent a significant advancement in targeted cancer treatment.
• Identifying robust SL pairs is critical for expanding these therapies, but experimental screening is limited by the vast number of gene combinations.

Purpose of the Study:

• To review computational methods for predicting synthetic lethality (SL) pairs.
• To summarize relevant data resources and highlight the application of various computational approaches.
• To discuss challenges and future directions in computational SL prediction for cancer therapy.

Main Methods:

• Review of existing literature on synthetic lethality (SL) screening and prediction.
• Summary of data resources and databases relevant to SL pair identification.
• Categorization and elaboration of computational prediction methods, including statistical, network-based, classical machine learning, and deep learning approaches, with a focus on negative sampling.

Main Results:

• Computational methods significantly complement experimental approaches by reducing the search space for SL pairs.
• Various computational strategies, from statistical to deep learning, have been applied to predict SL interactions.
• The review consolidates information on SL-related data, prediction tools, and methodologies.

Conclusions:

• Computational prediction is essential for efficient identification of synthetic lethality (SL) pairs, accelerating the development of targeted cancer therapies.
• Further development of computational models and integration of diverse data sources are needed to improve prediction accuracy.
• Addressing challenges in negative sampling and model interpretability will be key for future advancements in SL-based drug discovery.