Back

BLINK: Ultrafast tandem mass spectrometry cosine similarity scoring

Harwood, T. V.; Treen, D.; Wang, M.; de Jong, W.; Northen, T.; Bowen, B. P.

2023-02-15 bioinformatics
10.1101/2023.02.14.528550 bioRxiv
Show abstract

SummaryMetabolomics has a long history of using cosine similarity to match experimental tandem mass spectra to databases for compound identification. Here we introduce the Blur-and-Link (BLINK) approach for scoring cosine similarity. BLINK calculates substantially equivalent cosine similarity scores (>99% identification agreement) over 1000 times faster than commonly used loop-based implementations by bypassing fragment alignment and simultaneously scoring all pairs of spectra using sparse matrix operations. This performance improvement can enable calculations to be performed that would typically be limited by time and available computational resources. Availability and ImplementationBLINK is implemented in Python3 and is published under a modified open source license. Code and license are available on Github: https://github.com/biorack/blink

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.