Back

TIR-Learner v4: Accelerated annotation of terminal inverted repeat transposons

Gerhardt, K.; Ou, S.

2026-07-23 bioinformatics
10.64898/2026.07.21.739826 bioRxiv
Show abstract

The rapid expansion of high-quality, nearly complete eukaryotic genomes demands computational accelerations of existing bioinformatic infrastructure. TIR-Learner has been widely used for the de novo identification of Terminal Inverted Repeat (TIR) transposons, but suffers from slow runtime and a large memory footprint. Here we present TIR-Learner v4, a complete rewrite of TIR-Learner v3 and key supporting programs, which accelerates runtimes by two orders of magnitude and caps at a low RAM footprint irrespective of genome size. We demonstrate the scalability of TIR-Learner v4 by annotating TIRs in all 579 VGP Phase I genomes in around 4 hours.

Matching journals

The top 6 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.