Back

A deep learning model embedded framework to distinguish DNA and RNA mutations directly from RNA-seq

Fu, Z.-C.; Gao, B.-Q.; Nan, F.; Ma, X.-K.; Yang, L.

2023-10-20 bioinformatics
10.1101/2023.10.17.562625 bioRxiv
Show abstract

We develop a stepwise computational framework, called DEMINING, to directly detect expressed DNA and RNA mutations in RNA deep sequencing data. DEMINING incorporates a deep learning model named DeepDDR, which facilitates the separation of expressed DNA mutations from RNA mutations after RNA-seq read mapping and pileup. When applied in RNA-seq of acute myeloid leukemia patients, DEMINING uncovered previously-underappreciated DNA and RNA mutations, some associated with the upregulated expression of host genes or the production of neoantigens. Finally, we demonstrate that DEMINING could precisely classify DNA and RNA mutations in RNA-seq data from non-primate species through the utilization of transfer learning.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.