Accurate isoform quantification in bulk and single-cell RNA-seq by integrating multi-platform sequencing data and sequence-to-expression models

Multi-platform sequencing integration with MPAQT achieves state-of-the-art isoform quantification, validated by simulations and experimental benchmarks.

• Why it matters: Accurate isoform quantification is essential for understanding gene regulation and cellular diversity, but current methods face limitations in resolution and bias.
• What they did: The study developed MPAQT, a generative model combining short-read and long-read sequencing data, and applied machine learning to analyze transcript abundances in human embryonic stem cell differentiation.
• The result: Findings reveal UTRs as key determinants of isoform proportions and exon usage, mediated through sequence features interacting with RNA-binding proteins, advancing insights into transcriptomic complexity.