BaCoN (Balanced Correlation Network) improves prediction of gene buffering.

Area of Science:

• Genomics
• Systems Biology
• Computational Biology

Background:

• Gene buffering ensures cellular robustness by allowing gene compensation.
• Predicting genome-wide buffering is challenging due to experimental infeasibility.
• Gene buffering capacity is linked to expression levels and fitness phenotypes.

Purpose of the Study:

• To develop a computational method for predicting gene buffering relationships genome-wide.
• To identify potential gene buffering interactions within cancer cell lines.
• To evaluate the performance of the developed method against existing approaches.

Main Methods:

• Developed Balanced Correlation Network (BaCoN), an unsupervised correction method for expression-vs-fitness correlation networks.
• Associated CRISPR-Cas9 screening fitness effects with transcriptomic data across 1019 Cancer Dependency Map (DepMap) cell lines.
• Quantified 147 million potential buffering relationships.

Main Results:

• BaCoN outperformed state-of-the-art methods in predicting gene buffering.
• Identified 808 high-confidence buffering predictions.
• Found that buffering paralogs are typically located on different chromosomes.

Conclusions:

• BaCoN is an effective tool for predicting gene buffering from large-scale datasets like DepMap.
• The method's performance improves with increased screening data and gene coverage.
• The findings provide insights into the genomic organization of buffering gene pairs.