Reproduce “BIOTRAM: Biochemistry-informed Transformer for Relative Abundance via Metabarcoding” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Emma Boehly, Scott C. Lowe, Ken A. Thompson, Anthony Fuller, Will Armstrong, Iuliia Zarubiieva, Neelu Madan, Tiancheng Gao and 65 more
BIOTRAM corrects bulk metabarcoding read counts into estimated insect counts by modeling specimen traits and laboratory processes that affect read yield. It combines per-BIN yield estimates with bottle-level amplification and replicate information. On held-out bottles without excess specimens, it lowers count error to 0.699 from 0.723 and composition JSD to 0.1200 from 0.1234 against the best matched baseline, though all models degrade under distribution shift.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.