#39 · Transcriptomics

Transcriptome-wide resources and pipelines refine human RNA annotation and regulation

The dominant thread is computational infrastructure and reanalysis for transcriptome-scale data. A pan-transcriptome assembly yields a transcript-resolved human enhancer RNA catalogue of 36,536 transcripts, distributed as GTF and junction BED resources via eRNAkit. ORFannotate is a GTF-native Python tool that predicts ORFs from transcript models, reinserts exon-aware CDS and UTR features, and annotates Kozak sequences and uORFs. PARANOiD automates iCLIP and iCLIP2 read processing, including cross-link site calling and motif detection. On the biological side, a uniform reanalysis of 1,515 matched ribosome profiling and RNA-seq datasets from human and mouse examines translational buffering, and a model using RBP binding and sequence predicts neuron versus glia exon inclusion, with neuronal splicing reported as less sequence-dependent. Together these items connect expression, translation, and splicing regulation through reproducible, transcript-resolved analyses.