Publications
We seek to understand how alternative splicing and other post-transcriptional RNA processes shape gene expression, and how this mediates genetic variation to human disease. We develop statistical and computational methods, spanning single-cell to population-scale data, and apply them to uncover the regulatory logic of the transcriptome.
Below are highlights from our published and ongoing work.
Genetic and functional mapping of alternative splicing using LeafCutter2
LeafCutter2 enables identification and quantification of unproductive (NMD-associated) splicing from short-read RNA-seq data. Applying it across human tissues and non-human animals reveals a broad landscape of unproductive splicing, and analysis of data from the AD Functional Genomics (FunGen-xQTL) consortium uncovers novel associations between unproductive splicing and Alzheimer's disease risk.
Reference-free analysis of mRNA isoform variation in large-scale RNA-seq datasets
Torino combines a matrix factorization approach with splice-junction analysis to uncover the main structures underlying the transcriptome without relying on annotations. Torino finds thousands of novel transcript structures relevant to complex disease, including cryptic splicing, intron retention, and alternative polyadenylation.
· Preprint
Global impact of unproductive splicing on human gene expression
Alternative splicing leading to NMD-targeted mRNAs is widespread across the transcriptome. This unproductive splicing impacts gene expression levels and is an important link between genetic variants, eQTLs, and complex traits.
Coverage-dependent bias creates the appearance of binary splicing in single cells
Previous studies reported that single-cell alternative splicing is binary: most cells produce either one isoform or the other, but rarely both. We found that the low capture efficiency of scRNA-seq accounts for most of this pattern — in reality, single cells produce multiple isoforms from the same gene, but most are missed experimentally.
About my name: I am always grateful when other scientists cite my papers! If you do, please note that Buen Abad is part of a compound surname (not a middle name), so the correct citation format is Buen Abad Najar, C.F. Automatic reference managers often write it incorrectly into Najar, C.F.B.A. Please consider correcting this if you notice it. For informal purposes, you can abbreviate my name to Carlos Buen Abad.