Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312036251> ?p ?o ?g. }
- W4312036251 abstract "Abstract Over the past decade, neural networks have been successful at making predictions from biological sequences, especially in the context of regulatory genomics. As in other fields of deep learning, tools have been devised to extract features such as sequence motifs that can explain the predictions made by a trained network. Here we intend to go beyond explainable machine learning and introduce SEISM, a selective inference procedure to test the association between these extracted features and the predicted phenotype. In particular, we discuss how training a one-layer convolutional network is formally equivalent to selecting motifs maximizing some association score. We adapt existing sampling-based selective inference procedures by quantizing this selection over an infinite set to a large but finite grid. Finally, we show that sampling under a specific choice of parameters is sufficient to characterize the composite null hypothesis typically used for selective inference—a result that goes well beyond our particular framework. We illustrate the behavior of our method in terms of calibration, power and speed and discuss its power/speed trade-off with a simpler data-split strategy. SEISM paves the way to an easier analysis of neural networks used in regulatory genomics, and to more powerful methods for genome wide association studies (GWAS)." @default.
- W4312036251 created "2023-01-04" @default.
- W4312036251 creator A5050991089 @default.
- W4312036251 creator A5059163151 @default.
- W4312036251 creator A5070748025 @default.
- W4312036251 creator A5081394402 @default.
- W4312036251 date "2022-12-23" @default.
- W4312036251 modified "2023-10-18" @default.
- W4312036251 title "Neural Networks beyond explainability: Selective inference for sequence motifs" @default.
- W4312036251 cites W1019830208 @default.
- W4312036251 cites W1885924565 @default.
- W4312036251 cites W1983405741 @default.
- W4312036251 cites W2004119563 @default.
- W4312036251 cites W2026690931 @default.
- W4312036251 cites W2041877620 @default.
- W4312036251 cites W2046343410 @default.
- W4312036251 cites W2057253402 @default.
- W4312036251 cites W2103777723 @default.
- W4312036251 cites W2145091349 @default.
- W4312036251 cites W2173441045 @default.
- W4312036251 cites W2198606573 @default.
- W4312036251 cites W2247766769 @default.
- W4312036251 cites W2293040502 @default.
- W4312036251 cites W2724177157 @default.
- W4312036251 cites W2725988230 @default.
- W4312036251 cites W2767817150 @default.
- W4312036251 cites W2950417030 @default.
- W4312036251 cites W2952239877 @default.
- W4312036251 cites W2963417959 @default.
- W4312036251 cites W2995175313 @default.
- W4312036251 cites W3020975691 @default.
- W4312036251 cites W3099470970 @default.
- W4312036251 cites W3103643510 @default.
- W4312036251 cites W3130819465 @default.
- W4312036251 cites W3139265997 @default.
- W4312036251 cites W3157449380 @default.
- W4312036251 cites W3203588026 @default.
- W4312036251 cites W4225622462 @default.
- W4312036251 cites W4300862723 @default.
- W4312036251 doi "https://doi.org/10.1101/2022.12.23.521748" @default.
- W4312036251 hasPublicationYear "2022" @default.
- W4312036251 type Work @default.
- W4312036251 citedByCount "0" @default.
- W4312036251 crossrefType "posted-content" @default.
- W4312036251 hasAuthorship W4312036251A5050991089 @default.
- W4312036251 hasAuthorship W4312036251A5059163151 @default.
- W4312036251 hasAuthorship W4312036251A5070748025 @default.
- W4312036251 hasAuthorship W4312036251A5081394402 @default.
- W4312036251 hasBestOaLocation W43120362511 @default.
- W4312036251 hasConcept C105795698 @default.
- W4312036251 hasConcept C111472728 @default.
- W4312036251 hasConcept C119857082 @default.
- W4312036251 hasConcept C138885662 @default.
- W4312036251 hasConcept C142853389 @default.
- W4312036251 hasConcept C151730666 @default.
- W4312036251 hasConcept C154945302 @default.
- W4312036251 hasConcept C177264268 @default.
- W4312036251 hasConcept C191988596 @default.
- W4312036251 hasConcept C199360897 @default.
- W4312036251 hasConcept C2776214188 @default.
- W4312036251 hasConcept C2778112365 @default.
- W4312036251 hasConcept C2779343474 @default.
- W4312036251 hasConcept C33923547 @default.
- W4312036251 hasConcept C41008148 @default.
- W4312036251 hasConcept C50644808 @default.
- W4312036251 hasConcept C54355233 @default.
- W4312036251 hasConcept C81363708 @default.
- W4312036251 hasConcept C86803240 @default.
- W4312036251 hasConceptScore W4312036251C105795698 @default.
- W4312036251 hasConceptScore W4312036251C111472728 @default.
- W4312036251 hasConceptScore W4312036251C119857082 @default.
- W4312036251 hasConceptScore W4312036251C138885662 @default.
- W4312036251 hasConceptScore W4312036251C142853389 @default.
- W4312036251 hasConceptScore W4312036251C151730666 @default.
- W4312036251 hasConceptScore W4312036251C154945302 @default.
- W4312036251 hasConceptScore W4312036251C177264268 @default.
- W4312036251 hasConceptScore W4312036251C191988596 @default.
- W4312036251 hasConceptScore W4312036251C199360897 @default.
- W4312036251 hasConceptScore W4312036251C2776214188 @default.
- W4312036251 hasConceptScore W4312036251C2778112365 @default.
- W4312036251 hasConceptScore W4312036251C2779343474 @default.
- W4312036251 hasConceptScore W4312036251C33923547 @default.
- W4312036251 hasConceptScore W4312036251C41008148 @default.
- W4312036251 hasConceptScore W4312036251C50644808 @default.
- W4312036251 hasConceptScore W4312036251C54355233 @default.
- W4312036251 hasConceptScore W4312036251C81363708 @default.
- W4312036251 hasConceptScore W4312036251C86803240 @default.
- W4312036251 hasLocation W43120362511 @default.
- W4312036251 hasLocation W43120362512 @default.
- W4312036251 hasLocation W43120362513 @default.
- W4312036251 hasLocation W43120362514 @default.
- W4312036251 hasOpenAccess W4312036251 @default.
- W4312036251 hasPrimaryLocation W43120362511 @default.
- W4312036251 hasRelatedWork W2337926734 @default.
- W4312036251 hasRelatedWork W2978290780 @default.
- W4312036251 hasRelatedWork W3027997911 @default.
- W4312036251 hasRelatedWork W4287776258 @default.
- W4312036251 hasRelatedWork W4308353688 @default.
- W4312036251 hasRelatedWork W4312501200 @default.
- W4312036251 hasRelatedWork W4313050734 @default.