End-to-end sequence-structure-function meta-learning predicts genome-wide chemical-protein interactions for dark proteins.
Tian CaiLi XieShuo ZhangMuge ChenDi HeAmitesh BadkulYang LiuHari Krishna NamballaMichael DoroganWayne W. HardingCameron MuraPhilip E. BourneLei XiePublished in: PLoS Comput. Biol. (2023)
Keyphrases
- end to end
- genome wide
- meta learning
- protein protein interactions
- high throughput
- protein interaction
- dna binding
- biological data
- protein function
- inductive learning
- microarray
- sequence similarity
- computational methods
- biological processes
- learning tasks
- model selection
- functional modules
- systems biology
- protein protein interaction networks
- protein interaction networks
- protein complexes
- genomic data
- decision trees
- gene ontology
- machine learning algorithms
- biological networks
- machine learning
- amino acids
- biomedical literature
- ad hoc networks
- gene expression
- real time
- protein sequences
- transfer learning
- high speed
- reinforcement learning
- learning algorithm
- data mining
- data sets