Long Read Alignment Assembler (LRAA)

Isoform Discovery and/or Quantification from Long Read RNA-Seq

Visit the LRAA wiki for user documentation

Docker images

us-central1-docker.pkg.dev/methods-dev-lab/lraa/lraa:latest

:latest is the current official public release -- not the newest build. Use it, or pin the release version it resolves to if you need a fixed reference. The version it names moves only when a release is made.

Development builds are never published under :latest; they are tagged <version>-testing or <version>-<shortsha>. See Docker/README.md for the full tag policy and the build scripts that enforce it.

Fragmented and preassembled inputs

--pasa_mode chains fragmented transcript alignments (ESTs, partial cDNAs, staggered reads) that share introns into longer models; --preassembled_input accepts single-alignment support for preassembled transcripts (Trinity, StringTie). Long-read RNA-seq needs neither. See docs/pasa_mode.md.

Developer utilities

util/ascii_isoform_view.py draws transcript structures as stacked ASCII rows — reference GTF against LRAA.gtf, or a multipath path through the splice graph — with per-row structural verdicts. See docs/ascii_isoform_illustrator.md.