Our paper “Fast Mass Spectrometry Search and Clustering of Untargeted Metabolomics Data” is published in Nature Biotechnology on January 2, 2024. Our method runs on an indexing based algorithm for one versus all or all versus all fast similarity dot-product score calculation of mass spectrometry data, making our tools over two magnitudes faster than the state of the art. Additionally, our software enables searching and analysis across the whole GNPS (currently contains 717 million MS data) dataset using single CPU, a goal that could never be accomplished by existing methods.

Code availability

MASST+ and Networking+ are available at https://github.com/mohimanilab/MASSTplus. Other custom software used in this work includes Seq2Ripp, PepNovo and Dereplicator

Data Availability

The datasets analyzed are available at gnps.ucsd.edu. Accession codes related to the lanthipeptides part of the study are MSV000090476, MSV000090473, MSV000090472, MSV000090471, MSV000090457, MSV000089818, MSV000089817, MSV000089816, MSV000089815, MSV000089813, MSV000088816, MSV000088801, MSV000088800, MSV000088764 and MSV000088763. For comparing MASST+ and Networking+ against previous state-of-the-art tools, datasets MSV000078787, clustered GNPS, and unclustered GNPS were used. The accession codes for clustered GNPS and unclustered GNPS are available in Supplementary Data 1.