Integrative discovery and functional dependency mapping uncover essential chimeric RNAs in the active transcriptome

Chimeric RNAs are a class of molecules defined as RNA transcripts containing genetic information for multiple genes. While their expression patterns across tissue types and disease states have been well documented, the biological function of these transcripts remains poorly understood. In this study, we developed an integrative prediction pipeline to establish an atlas of 569,640 chimeric RNAs across 1019 cancer cell lines. Leveraging this resource, we performed a systematic functional dependency screen by integrating our predictions with perturbation datasets from The Cancer Dependency Map project. This approach moves beyond prediction to identifying high-confidence functional chimeric RNAs essential for cellular processes, including a subset of recurrent chimeric transcripts generated by cis - splicing of adjacent genes. We demonstrate that depletion of selected candidate chimeras disrupts cell cycle progression and cellular fitness and that these transcripts are frequently dysregulated in primary breast tumors. Together, these results demonstrate that chimeric RNAs constitute a previously underappreciated layer of gene expression, expanding the functional genome beyond conventional genetics.