FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4360821468> ?p ?o ?g. }

• W4360821468 abstract "Abstract The most serious drawback underlying the biological annotation of Whole Genome Sequence data is the p>>n problem, meaning that the number of polymorphic variants (p) is much larger than the number of available phenotypic records (n). Therefore, the major aim of the study was to propose a way to circumvent the problem by combining a LASSO logistic regression model with Deep Learning (DL). That was illustrated by a practical biological problem of classification of cows into mastitis-susceptible or mastitis-resistant, based on genotypes of Single Nucleotide Polymorphisms (SNPs) identified in their WGS. Among several DL architectures proposed via optimisation of DL hyperparameters using the Optuna software, imposed on different SNP sub-sets defined by LASSO logistic regressions with different penalty values, the architecture with 204,642 SNPs was selected as the best one. This architecture was composed of 2 layers with respectively 7 and 46 units per layer as well as respective drop-out rates of 0.210 and 0.358. The classification of the test data set resulted in the AUC=0.750, accuracy=0.650, sensitivity=0.600, and specificity=0.700 was selected as the best model and thus proceeded to genomic and functional annotations. Significant SNPs were selected based on the SHapley Additive exPlanation values transformed to Z-scores to assess the underlying type I-error. These SNPs were annotated to genes. As a final result, a single GO term related to the biological process and thirteen GO terms related to the molecular function were significantly enriched in the gene set that corresponded to the significant SNPs. Author Summary Our objective is to distinguish between cows that are susceptible and resistant to bovine mastitis by analysing their genomic data. However, we face a significant challenge due to the large number of single nucleotide polymorphisms (SNPs) and limited sample size. To address this challenge, we utilize two methods: feature selection algorithms and deep learning. We experiment with various ways of implementing these techniques and evaluate their performance on a validation set. Our findings reveal that the optimal approach can accurately predict a cow’s susceptibility or resistance status around 65% of the time. Additionally, we employ a technique to identify the most crucial SNPs and their biological functions. Our results indicate that some of these SNPs are related to immune response or protein synthesis pathways, implying that they may affect the cow’s health and productivity." @default.
• W4360821468 created "2023-03-25" @default.
• W4360821468 creator A5033758299 @default.
• W4360821468 creator A5040897798 @default.
• W4360821468 creator A5041277461 @default.
• W4360821468 creator A5049061860 @default.
• W4360821468 creator A5058863957 @default.
• W4360821468 creator A5060834608 @default.
• W4360821468 creator A5081824623 @default.
• W4360821468 creator A5090006923 @default.
• W4360821468 date "2023-03-24" @default.
• W4360821468 modified "2023-10-16" @default.
• W4360821468 title "An explainable deep learning classifier of bovine mastitis based on whole genome sequence data - circumventing the p>>>n problem" @default.
• W4360821468 cites W1501531009 @default.
• W4360821468 cites W1965092590 @default.
• W4360821468 cites W1974890191 @default.
• W4360821468 cites W1992643433 @default.
• W4360821468 cites W1996260257 @default.
• W4360821468 cites W1997073431 @default.
• W4360821468 cites W2007739294 @default.
• W4360821468 cites W2037338504 @default.
• W4360821468 cites W2039139862 @default.
• W4360821468 cites W2043235003 @default.
• W4360821468 cites W2070213312 @default.
• W4360821468 cites W2102619694 @default.
• W4360821468 cites W2103017472 @default.
• W4360821468 cites W2103441770 @default.
• W4360821468 cites W2108234281 @default.
• W4360821468 cites W2113499361 @default.
• W4360821468 cites W2119180969 @default.
• W4360821468 cites W2130395838 @default.
• W4360821468 cites W2131271579 @default.
• W4360821468 cites W2135046866 @default.
• W4360821468 cites W2137959503 @default.
• W4360821468 cites W2141545183 @default.
• W4360821468 cites W2149992227 @default.
• W4360821468 cites W2169628503 @default.
• W4360821468 cites W2177394834 @default.
• W4360821468 cites W2275092079 @default.
• W4360821468 cites W2293418890 @default.
• W4360821468 cites W2417483443 @default.
• W4360821468 cites W2489962810 @default.
• W4360821468 cites W2733861029 @default.
• W4360821468 cites W2748517095 @default.
• W4360821468 cites W2781815607 @default.
• W4360821468 cites W2789367970 @default.
• W4360821468 cites W2789828535 @default.
• W4360821468 cites W2790283876 @default.
• W4360821468 cites W2794803511 @default.
• W4360821468 cites W2799274693 @default.
• W4360821468 cites W2802639922 @default.
• W4360821468 cites W2883972171 @default.
• W4360821468 cites W2902196707 @default.
• W4360821468 cites W2914884022 @default.
• W4360821468 cites W2949676527 @default.
• W4360821468 cites W2971969289 @default.
• W4360821468 cites W3036779511 @default.
• W4360821468 cites W3092916076 @default.
• W4360821468 cites W3176838620 @default.
• W4360821468 cites W3181530525 @default.
• W4360821468 cites W3211402536 @default.
• W4360821468 cites W4200579482 @default.
• W4360821468 cites W4206945683 @default.
• W4360821468 cites W4220831425 @default.
• W4360821468 cites W4221101236 @default.
• W4360821468 cites W4225308588 @default.
• W4360821468 cites W4225597912 @default.
• W4360821468 cites W4244197312 @default.
• W4360821468 cites W4283362956 @default.
• W4360821468 cites W4285744259 @default.
• W4360821468 cites W4289274885 @default.
• W4360821468 cites W4294216483 @default.
• W4360821468 cites W4307122273 @default.
• W4360821468 doi "https://doi.org/10.1101/2023.03.23.533903" @default.
• W4360821468 hasPublicationYear "2023" @default.
• W4360821468 type Work @default.
• W4360821468 citedByCount "0" @default.
• W4360821468 crossrefType "posted-content" @default.
• W4360821468 hasAuthorship W4360821468A5033758299 @default.
• W4360821468 hasAuthorship W4360821468A5040897798 @default.
• W4360821468 hasAuthorship W4360821468A5041277461 @default.
• W4360821468 hasAuthorship W4360821468A5049061860 @default.
• W4360821468 hasAuthorship W4360821468A5058863957 @default.
• W4360821468 hasAuthorship W4360821468A5060834608 @default.
• W4360821468 hasAuthorship W4360821468A5081824623 @default.
• W4360821468 hasAuthorship W4360821468A5090006923 @default.
• W4360821468 hasBestOaLocation W43608214681 @default.
• W4360821468 hasConcept C104317684 @default.
• W4360821468 hasConcept C119857082 @default.
• W4360821468 hasConcept C135763542 @default.
• W4360821468 hasConcept C136764020 @default.
• W4360821468 hasConcept C141231307 @default.
• W4360821468 hasConcept C151956035 @default.
• W4360821468 hasConcept C153209595 @default.
• W4360821468 hasConcept C154945302 @default.
• W4360821468 hasConcept C24432333 @default.
• W4360821468 hasConcept C33923547 @default.
• W4360821468 hasConcept C37616216 @default.
• W4360821468 hasConcept C41008148 @default.
• W4360821468 hasConcept C54355233 @default.

• next