Matches in SemOpenAlex for { <https://semopenalex.org/work/W4283787644> ?p ?o ?g. }
- W4283787644 abstract "ABSTRACT Taking sequences as the only inputs, the class of de novo deep learning (DL) models for RNA secondary structure prediction has achieved far superior performances than traditional algorithms. However, key questions remain over the statistical underpinning of such models that make no use of physical laws or co-evolutionary information. We present a quantitative study of the capacity and generalizability of a series of de novo DL models, with a minimal two-module architecture and no post-processing, under varied distributions of the seen and unseen sequences. Our DL models outperform existing methods on commonly used benchmark datasets and demonstrate excellent learning capacities under all sequence distributions. These DL models generalize well over non-identical unseen sequences, but the generalizability degrades rapidly as the sequence distributions of the seen and unseen datasets become dissimilar. Examinations of RNA family-specific behaviors manifest not only disparate familydependent performances but substantial generalization gaps within the same family. We further determine how model generalization decreases with the decrease of sequence similarity via pairwise sequence alignment, providing quantitative insights into the limitations of statistical learning. Model generalizability thus poses a major hurdle for practical uses of de novo DL models and several tenable avenues for future advances are discussed." @default.
- W4283787644 created "2022-07-04" @default.
- W4283787644 creator A5057146487 @default.
- W4283787644 date "2022-07-02" @default.
- W4283787644 modified "2023-09-25" @default.
- W4283787644 title "Decisive Roles of Sequence Distributions in the Generalizability of <i>de novo</i> Deep Learning Models for RNA Secondary Structure Prediction" @default.
- W4283787644 cites W166206240 @default.
- W4283787644 cites W1965243792 @default.
- W4283787644 cites W1981576666 @default.
- W4283787644 cites W1984867418 @default.
- W4283787644 cites W1987592711 @default.
- W4283787644 cites W2003993220 @default.
- W4283787644 cites W2020546070 @default.
- W4283787644 cites W2025763720 @default.
- W4283787644 cites W2026181790 @default.
- W4283787644 cites W2027141824 @default.
- W4283787644 cites W2038598202 @default.
- W4283787644 cites W2068205741 @default.
- W4283787644 cites W2071883352 @default.
- W4283787644 cites W2086561953 @default.
- W4283787644 cites W2097924797 @default.
- W4283787644 cites W2098571862 @default.
- W4283787644 cites W2102017611 @default.
- W4283787644 cites W2106478835 @default.
- W4283787644 cites W2121918723 @default.
- W4283787644 cites W2133839648 @default.
- W4283787644 cites W2141157874 @default.
- W4283787644 cites W2142038007 @default.
- W4283787644 cites W2142678478 @default.
- W4283787644 cites W2145149786 @default.
- W4283787644 cites W2155858493 @default.
- W4283787644 cites W2159548583 @default.
- W4283787644 cites W2170747616 @default.
- W4283787644 cites W2342878844 @default.
- W4283787644 cites W2414122243 @default.
- W4283787644 cites W2536860838 @default.
- W4283787644 cites W2554065905 @default.
- W4283787644 cites W2596417560 @default.
- W4283787644 cites W2725568163 @default.
- W4283787644 cites W2759571676 @default.
- W4283787644 cites W2808896479 @default.
- W4283787644 cites W2866340037 @default.
- W4283787644 cites W2910705748 @default.
- W4283787644 cites W2919479895 @default.
- W4283787644 cites W2933083636 @default.
- W4283787644 cites W2945901478 @default.
- W4283787644 cites W2951298881 @default.
- W4283787644 cites W2954102902 @default.
- W4283787644 cites W2983889514 @default.
- W4283787644 cites W2990528340 @default.
- W4283787644 cites W2998238598 @default.
- W4283787644 cites W3032032947 @default.
- W4283787644 cites W3047090055 @default.
- W4283787644 cites W3087202564 @default.
- W4283787644 cites W3111119989 @default.
- W4283787644 cites W3126773939 @default.
- W4283787644 cites W3127238141 @default.
- W4283787644 cites W3161531818 @default.
- W4283787644 cites W3163993681 @default.
- W4283787644 cites W3177828909 @default.
- W4283787644 cites W3212533323 @default.
- W4283787644 cites W4205788935 @default.
- W4283787644 cites W4210378368 @default.
- W4283787644 cites W4220961504 @default.
- W4283787644 cites W4280513900 @default.
- W4283787644 doi "https://doi.org/10.1101/2022.06.29.498185" @default.
- W4283787644 hasPublicationYear "2022" @default.
- W4283787644 type Work @default.
- W4283787644 citedByCount "1" @default.
- W4283787644 countsByYear W42837876442023 @default.
- W4283787644 crossrefType "posted-content" @default.
- W4283787644 hasAuthorship W4283787644A5057146487 @default.
- W4283787644 hasBestOaLocation W42837876441 @default.
- W4283787644 hasConcept C103278499 @default.
- W4283787644 hasConcept C105795698 @default.
- W4283787644 hasConcept C108583219 @default.
- W4283787644 hasConcept C114289077 @default.
- W4283787644 hasConcept C115961682 @default.
- W4283787644 hasConcept C119857082 @default.
- W4283787644 hasConcept C13280743 @default.
- W4283787644 hasConcept C134306372 @default.
- W4283787644 hasConcept C154945302 @default.
- W4283787644 hasConcept C177148314 @default.
- W4283787644 hasConcept C184898388 @default.
- W4283787644 hasConcept C185798385 @default.
- W4283787644 hasConcept C205649164 @default.
- W4283787644 hasConcept C27158222 @default.
- W4283787644 hasConcept C2778112365 @default.
- W4283787644 hasConcept C33923547 @default.
- W4283787644 hasConcept C41008148 @default.
- W4283787644 hasConcept C54355233 @default.
- W4283787644 hasConcept C70721500 @default.
- W4283787644 hasConcept C86803240 @default.
- W4283787644 hasConceptScore W4283787644C103278499 @default.
- W4283787644 hasConceptScore W4283787644C105795698 @default.
- W4283787644 hasConceptScore W4283787644C108583219 @default.
- W4283787644 hasConceptScore W4283787644C114289077 @default.
- W4283787644 hasConceptScore W4283787644C115961682 @default.
- W4283787644 hasConceptScore W4283787644C119857082 @default.
- W4283787644 hasConceptScore W4283787644C13280743 @default.