Matches in SemOpenAlex for { <https://semopenalex.org/work/W4308398768> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4308398768 abstract "<h3>Background</h3> To determine the specificity of T cells based on their receptor sequence is a demanding task due to cross-reactivity, complicated patterns and limited size of public dataset. An effective computational model, which finds CDR3 patterns of shared antigen specificity in the existing dataset, can predict specificity of T cells accurately. In this work, we developed a deep learning methodology that computes the similarity among T cells in terms of antigen specificity using <i>k</i>-mer features. <h3>Methods</h3> Our model consists of two parts. First, it encodes every overlapping <i>k</i>-mers of CDR3 into numerical vectors. We parallelize such <i>k</i>-mer encodings into several allowable ways, so that the independent semantics of each <i>k</i>-mers are effectively learned. Second, among the encoded <i>k</i>-mer features, we select only meaningful <i>k-</i>mers using a self-attention structure. By doing this, we remove unwanted correlations among overlapping <i>k</i>-mers. We train our model with preprocessed public datasets: IEDB, VDJdb and McPAS. We optimize the overall process to find an optimal contrastive predictive coding, which is an unsupervised objective function. After optimization, we define a kernel function of <i>k</i>-mer features to define similarity between two CDR3s. <h3>Results</h3> We designed an one-of-many unsupervised task: for a given arbitrary CDR3 sequence, whether our model can correctly select CDR3 with a similar specificity among N randomly sampled candidates. With N=10, our model achieves accuracy 0.3 for an independent dataset. We also test supervised task: whether our model can induce probable cognate antigens for a given CDR3. Our model achieves precision 0.7. <h3>Conclusions</h3> Our deep learning model can extract <i>k</i>-mer information that only represents antigen specificity. This information is an invaluable numerical vector for computing similarity of antigen specificity. By doing this, our model can solve the one-of-many problem and predict the antigen specificity. In the future, our model will improve its performance as a size of training dataset grows." @default.
- W4308398768 created "2022-11-11" @default.
- W4308398768 creator A5003283147 @default.
- W4308398768 creator A5003733262 @default.
- W4308398768 creator A5003948286 @default.
- W4308398768 creator A5015198444 @default.
- W4308398768 creator A5023421086 @default.
- W4308398768 creator A5043014594 @default.
- W4308398768 creator A5046570826 @default.
- W4308398768 creator A5080009425 @default.
- W4308398768 date "2022-11-01" @default.
- W4308398768 modified "2023-10-06" @default.
- W4308398768 title "74 DeepTCRMatch: An effective way of computing T cells antigen specificity" @default.
- W4308398768 doi "https://doi.org/10.1136/jitc-2022-sitc2022.0074" @default.
- W4308398768 hasPublicationYear "2022" @default.
- W4308398768 type Work @default.
- W4308398768 citedByCount "0" @default.
- W4308398768 crossrefType "proceedings-article" @default.
- W4308398768 hasAuthorship W4308398768A5003283147 @default.
- W4308398768 hasAuthorship W4308398768A5003733262 @default.
- W4308398768 hasAuthorship W4308398768A5003948286 @default.
- W4308398768 hasAuthorship W4308398768A5015198444 @default.
- W4308398768 hasAuthorship W4308398768A5023421086 @default.
- W4308398768 hasAuthorship W4308398768A5043014594 @default.
- W4308398768 hasAuthorship W4308398768A5046570826 @default.
- W4308398768 hasAuthorship W4308398768A5080009425 @default.
- W4308398768 hasBestOaLocation W43083987681 @default.
- W4308398768 hasConcept C103278499 @default.
- W4308398768 hasConcept C114614502 @default.
- W4308398768 hasConcept C115961682 @default.
- W4308398768 hasConcept C14036430 @default.
- W4308398768 hasConcept C153180895 @default.
- W4308398768 hasConcept C154945302 @default.
- W4308398768 hasConcept C162324750 @default.
- W4308398768 hasConcept C187736073 @default.
- W4308398768 hasConcept C2778112365 @default.
- W4308398768 hasConcept C2780451532 @default.
- W4308398768 hasConcept C33923547 @default.
- W4308398768 hasConcept C41008148 @default.
- W4308398768 hasConcept C54355233 @default.
- W4308398768 hasConcept C74193536 @default.
- W4308398768 hasConcept C86803240 @default.
- W4308398768 hasConceptScore W4308398768C103278499 @default.
- W4308398768 hasConceptScore W4308398768C114614502 @default.
- W4308398768 hasConceptScore W4308398768C115961682 @default.
- W4308398768 hasConceptScore W4308398768C14036430 @default.
- W4308398768 hasConceptScore W4308398768C153180895 @default.
- W4308398768 hasConceptScore W4308398768C154945302 @default.
- W4308398768 hasConceptScore W4308398768C162324750 @default.
- W4308398768 hasConceptScore W4308398768C187736073 @default.
- W4308398768 hasConceptScore W4308398768C2778112365 @default.
- W4308398768 hasConceptScore W4308398768C2780451532 @default.
- W4308398768 hasConceptScore W4308398768C33923547 @default.
- W4308398768 hasConceptScore W4308398768C41008148 @default.
- W4308398768 hasConceptScore W4308398768C54355233 @default.
- W4308398768 hasConceptScore W4308398768C74193536 @default.
- W4308398768 hasConceptScore W4308398768C86803240 @default.
- W4308398768 hasLocation W43083987681 @default.
- W4308398768 hasOpenAccess W4308398768 @default.
- W4308398768 hasPrimaryLocation W43083987681 @default.
- W4308398768 hasRelatedWork W2084460089 @default.
- W4308398768 hasRelatedWork W2103444992 @default.
- W4308398768 hasRelatedWork W2110459882 @default.
- W4308398768 hasRelatedWork W2118043379 @default.
- W4308398768 hasRelatedWork W2141018987 @default.
- W4308398768 hasRelatedWork W2151022383 @default.
- W4308398768 hasRelatedWork W2603933437 @default.
- W4308398768 hasRelatedWork W2900794075 @default.
- W4308398768 hasRelatedWork W3183633970 @default.
- W4308398768 hasRelatedWork W3208309985 @default.
- W4308398768 isParatext "false" @default.
- W4308398768 isRetracted "false" @default.
- W4308398768 workType "article" @default.