Gram-LocEN: Interpretable prediction of subcellular multi-localization of Gram-positive and Gram-negative bacterial proteins

摘要: Abstract Bacteria have a highly organized internal architecture at the cellular level. Identifying subcellular localization of bacterial proteins is vital to infer their functions and design antibacterial drugs. Recent decades witnessed remarkable progress in protein by computational approaches. However, existing approaches following disadvantages: (1) prediction results are hard interpret; (2) they ignore multi-location proteins. To tackle these problems, this paper proposes an interpretable multi-label predictor, namely Gram-LocEN, for predicting both single- Gram-positive or Gram negative bacteria. By using elastic-net (EN) classifier, Gram-LocEN capable selecting location-specific essential features which play key roles determining localization. With features, not only where resides can be decided, but also why it locates there revealed. Experimental on two stringent benchmark datasets suggest that significantly outperforms state-of-the-art predictors Gram-negative For readers' convenience, web-server available http://bioinfo.eie.polyu.edu.hk/Gram-LocEN/ .