PlantCAD¶
Description¶
PlantCaduceus, with its short name of PlantCAD, is a plant DNA LM based on the Caduceus architecture, which extends the efficient Mamba linear-time sequence modeling framework to incorporate bi-directionality and reverse complement equivariance, specifically designed for DNA sequences. PlantCAD is pre-trained on a curated dataset of 16 Angiosperm genomes. PlantCAD showed state-of-the-art cross species performance in predicting TIS, TTS, Splice Donor and Splice Acceptor. The zero-shot of PlantCAD enables identifying genome-wide deleterious mutations and known causal variants in Arabidopsis, Sorghum and Maize.
Environment Modules¶
Run module spider plantcad to find out what environment modules are available for this application.
Environment Variables¶
- HPC_PLANTCAD_DIR - installation directory
- HPC_PLANTCAD_BIN - executable directory
- PLANTCADUCEUS_DIR - PlantCaduceus directory
Categories¶
biology, genomics