Multienzyme deep learning models improve peptide de novo sequencing by mass spectrometry proteomics
Gueto-Tettay, Carlos LU ; Tang, Di LU
; Happonen, Lotta
LU
; Heusel, Moritz
LU
; Khakzad, Hamed
; Malmström, Johan
LU
and Malmström, Lars
LU
(2023)
In PLoS Computational Biology
19(1).
- Abstract
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 systemcally varying the composition and size of the training set. We assessed the generated models' performances using two test sets composed of... (More)
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 systemcally 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 set's 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). MEMs 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 MEMs' proven improvements to de novo analysis will positively impact several applications, such as analyzing samples of high complexity, unknown nature, or the peptidomics field.
(Less)
- author
- Gueto-Tettay, Carlos
LU
; Tang, Di
LU
; Happonen, Lotta
LU
; Heusel, Moritz
LU
; Khakzad, Hamed
; Malmström, Johan
LU
and Malmström, Lars
LU
- organization
- publishing date
- 2023-01
- type
- Contribution to journal
- publication status
- published
- subject
- in
- PLoS Computational Biology
- volume
- 19
- issue
- 1
- article number
- e1010457
- publisher
- Public Library of Science (PLoS)
- external identifiers
-
- scopus:85147040937
- pmid:36668672
- ISSN
- 1553-734X
- DOI
- 10.1371/journal.pcbi.1010457
- project
- Properties of Protective Antibody Responses against Bacterial Pathogens
- language
- English
- LU publication?
- yes
- id
- 4c4d0359-e5be-4657-8c26-f47af388e476
- date added to LUP
- 2023-02-13 11:22:45
- date last changed
- 2024-11-15 16:40:53
@article{4c4d0359-e5be-4657-8c26-f47af388e476,
abstract = {{<p>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 systemcally 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 set's 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). MEMs 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 MEMs' proven improvements to de novo analysis will positively impact several applications, such as analyzing samples of high complexity, unknown nature, or the peptidomics field.</p>}},
author = {{Gueto-Tettay, Carlos and Tang, Di and Happonen, Lotta and Heusel, Moritz and Khakzad, Hamed and Malmström, Johan and Malmström, Lars}},
issn = {{1553-734X}},
language = {{eng}},
number = {{1}},
publisher = {{Public Library of Science (PLoS)}},
series = {{PLoS Computational Biology}},
title = {{Multienzyme deep learning models improve peptide de novo sequencing by mass spectrometry proteomics}},
url = {{http://dx.doi.org/10.1371/journal.pcbi.1010457}},
doi = {{10.1371/journal.pcbi.1010457}},
volume = {{19}},
year = {{2023}},
}