Improved protein contact prediction using dimensional hybrid residual networks and singularity enhanced loss function.
Yunda SiChengfei YanPublished in: Briefings Bioinform. (2021)
Keyphrases
- loss function
- subcellular localization
- polynomial neural networks
- support vector
- pairwise
- contact map
- protein interaction networks
- protein structure prediction
- logistic regression
- protein interaction
- prediction accuracy
- learning to rank
- empirical risk
- hinge loss
- risk minimization
- functional modules
- boosting framework
- reproducing kernel hilbert space
- protein structure
- amino acids
- convex loss functions
- complex networks
- stochastic gradient descent
- boosting algorithms
- prediction error
- multi class
- markov random field
- protein function
- protein sequences
- biological networks