SpectruMS: Foundation Model for Mass Spectrometry

AI training is similar to file compression (think .zip). Both tools make large files smaller. The added benefit of AI that it makes the compressed information more easily searchable.

SpectruMS is an attempt of Pagea Bio to compress all of GNPS, Massbank, PubChem and more into a single AI model. SpectruMS compressed petabytes of chemical data into a single, small, BART model:

I was there to design virtually every aspect of SpectruMS, from data curation, to AI training, to babysitting the TPU pod to deployment and serving on AWS.