A neural network pipeline ranks plausible multistep metabolic pathways after training binary classifiers on real versus artificially generated reactions from public databases and enzymatic templates.
New approach for understanding genome variations in kegg.Nucleic acids research, 47(D1):D590– D595
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Computational framework for multistep metabolic pathway design
A neural network pipeline ranks plausible multistep metabolic pathways after training binary classifiers on real versus artificially generated reactions from public databases and enzymatic templates.