1− 3 Fueled by progress in instrumentation over the previousĭecade, modern MS experiments can consist of millions of mass spectraĪnd require tens to hundreds of gigabytes storage space. To analyze the protein composition of biological samples and study Is currently the dominant analytical technique Based on experiments using various mass spectrometryĭata sets, HyperSpec produces results with comparable clustering qualityĪs state-of-the-art spectral clustering tools while achieving speedupsīy orders of magnitude, shortening the clustering runtime of overĢ1 million spectra from 4 h to only 24 min. ![]() The spectrum preprocessing time, which is a critical bottleneck during HyperSpec includes optimized data preprocessing modules to reduce Making it possible to run HyperSpec on graphics processing units toĪchieve extremely efficient spectral clustering performance. High parallelism that can be optimized using low-level hardware architectures, ![]() HDC shows promising clusteringĬapability while only requiring lightweight binary operations with This work, we present a fast spectral clustering tool, HyperSpec,īased on hyperdimensional computing (HDC). Optimal runtimes, this simply moves the processing bottleneck. However,īecause state-of-the-art spectral clustering tools fail to achieve Merging highly similar spectra to minimize data redundancy. Is an effective approach to speed up downstream data processing by Volumes has become progressively more challenging. Of mass spectrometry data per hour, processing these massive data As current shotgun proteomics experiments can produce
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |