Skip to content

Latest commit

 

History

History
13 lines (9 loc) · 677 Bytes

File metadata and controls

13 lines (9 loc) · 677 Bytes

ml-bcell-epitope

script1.py - Fully connected neural network barely beats SVM (~60% accuracy)
script2.py - Convolutional neural network with hot vector representation (~39% accuracy)
script3.py - Convolutional neural network with continuous distributed representation (TBD)

Paper Reference

Asgari, Ehsaneddin, and Mohammad RK Mofrad. "Continuous distributed representation of biological sequences for deep proteomics and genomics." PloS one 10.11 (2015): e0141287.

Dataset Reference

Gao J, Faraggi E, Zhou Y, Ruan J, Kurgan L. BEST: improved prediction of B-cell epitopes from antigen sequences. PLoS One. 2012; 7(6):e40104

http://biomine.ece.ualberta.ca/BEST/