Gueto Tettay, C. A. | Search bioRxiv

bioRxiv2024

Inferring the composition of the blood plasma proteome by a human proteome distribution atlas

The plasma proteome is maintained by the influx and efflux of proteins from surrounding organs and cells. To quantify the extent different organs and cells contribute to the plasma proteome composition, we developed a mass spectrometry-based proteomics strategy to infer the origin of proteins detected in human plasma in health and disease. First, we constructed an extensive human proteome atlas from 18 vascularized organs and the most abundant cell types in blood. Second, the atlas was interfaced with previous RNA/protein atlases to objectively define proteome wide protein-organ associations to enable both the inference of origin and the reproducible quantification of organ-specific proteins in plasma. We demonstrate that the resource can determine disease-specific quantitative changes of organ-enriched protein panels in three separate patient cohorts with infection, pancreatitis, and myocardial injury. The strategy can be extended to other diseases to advance our understanding of the processes contributing to plasma proteome dynamics.

systems biology↗

bioRxiv2022

Multienzyme deep learning models improve peptide de novo sequencing by mass spectrometry proteomics

Generating and analyzing overlapping peptides through multienzymatic digestion is an efficient procedure for de novo protein using from bottom-up mass spectrometry (MS). Despite improved instrumentation and software, de novo MS data analysis remains challenging. In recent years, deep learning models have represented a performance breakthrough. Incorporating that technology into de novo protein sequencing workflows require machine-learning models capable of handling highly diverse MS data. In this study, we analyzed the requirements for assembling such generalizable deep learning models by systematically varying the composition and size of the training set. We assessed the generated models performances using two test sets composed of peptides originating from the multienzyme digestion of samples from various species. The peptide recall values on the test sets showed that the deep learning models generated from a collection of highly N- and C-termini diverse peptides generalized 76% more over the termini-restricted ones. Moreover, expanding the training sets size by adding peptides from the multienzymatic digestion with five proteases of several species samples led to a 2-3 fold generalizability gain. Furthermore, we tested the applicability of these multienzyme deep learning (MEM) models by fully de novo sequencing the heavy and light monomeric chains of five commercial antibodies (mAbs). MEM models extracted over 10000 matching and overlapped peptides across six different proteases mAb samples, achieving a 100% sequence coverage for 8 of the ten polypeptide chains. We foretell that the MEM models proven improvements to de novo analysis will positively impact several applications, such as analyzing samples of high complexity, unknown nature, or the peptidomics field.

bioinformatics↗