In:
PLOS Computational Biology, Public Library of Science (PLoS), Vol. 18, No. 9 ( 2022-9-29), p. e1010561-
Abstract:
Selection protocols such as SELEX, where molecules are selected over multiple rounds for their ability to bind to a target of interest, are popular methods for obtaining binders for diagnostic and therapeutic purposes. We show that Restricted Boltzmann Machines (RBMs), an unsupervised two-layer neural network architecture, can successfully be trained on sequence ensembles from single rounds of SELEX experiments for thrombin aptamers. RBMs assign scores to sequences that can be directly related to their fitnesses estimated through experimental enrichment ratios. Hence, RBMs trained from sequence data at a given round can be used to predict the effects of selection at later rounds. Moreover, the parameters of the trained RBMs are interpretable and identify functional features contributing most to sequence fitness. To exploit the generative capabilities of RBMs, we introduce two different training protocols: one taking into account sequence counts, capable of identifying the few best binders, and another based on unique sequences only, generating more diverse binders. We then use RBMs model to generate novel aptamers with putative disruptive mutations or good binding properties, and validate the generated sequences with gel shift assay experiments. Finally, we compare the RBM’s performance with different supervised learning approaches that include random forests and several deep neural network architectures.
Type of Medium:
Online Resource
ISSN:
1553-7358
DOI:
10.1371/journal.pcbi.1010561
DOI:
10.1371/journal.pcbi.1010561.g001
DOI:
10.1371/journal.pcbi.1010561.g002
DOI:
10.1371/journal.pcbi.1010561.g003
DOI:
10.1371/journal.pcbi.1010561.g004
DOI:
10.1371/journal.pcbi.1010561.g005
DOI:
10.1371/journal.pcbi.1010561.g006
DOI:
10.1371/journal.pcbi.1010561.g007
DOI:
10.1371/journal.pcbi.1010561.g008
DOI:
10.1371/journal.pcbi.1010561.g009
DOI:
10.1371/journal.pcbi.1010561.t001
DOI:
10.1371/journal.pcbi.1010561.s001
DOI:
10.1371/journal.pcbi.1010561.s002
DOI:
10.1371/journal.pcbi.1010561.s003
DOI:
10.1371/journal.pcbi.1010561.s004
DOI:
10.1371/journal.pcbi.1010561.s005
DOI:
10.1371/journal.pcbi.1010561.s006
DOI:
10.1371/journal.pcbi.1010561.s007
DOI:
10.1371/journal.pcbi.1010561.s008
DOI:
10.1371/journal.pcbi.1010561.s009
DOI:
10.1371/journal.pcbi.1010561.s010
DOI:
10.1371/journal.pcbi.1010561.s011
DOI:
10.1371/journal.pcbi.1010561.s012
DOI:
10.1371/journal.pcbi.1010561.s013
DOI:
10.1371/journal.pcbi.1010561.s014
DOI:
10.1371/journal.pcbi.1010561.s015
DOI:
10.1371/journal.pcbi.1010561.s016
DOI:
10.1371/journal.pcbi.1010561.s017
DOI:
10.1371/journal.pcbi.1010561.s018
DOI:
10.1371/journal.pcbi.1010561.s019
DOI:
10.1371/journal.pcbi.1010561.s020
DOI:
10.1371/journal.pcbi.1010561.s021
DOI:
10.1371/journal.pcbi.1010561.s022
DOI:
10.1371/journal.pcbi.1010561.s023
DOI:
10.1371/journal.pcbi.1010561.s024
DOI:
10.1371/journal.pcbi.1010561.s025
DOI:
10.1371/journal.pcbi.1010561.s026
DOI:
10.1371/journal.pcbi.1010561.s027
Language:
English
Publisher:
Public Library of Science (PLoS)
Publication Date:
2022
detail.hit.zdb_id:
2193340-6
Permalink