PVPred-SCM: Improved Prediction and Analysis of Phage Virion Proteins Using Scoring Card Method

In this work, we developed a novel PVP predictor named PVPred-SCM which is an efficient and interpretable method for predicting and analyzing PVPs.

The same benchmark and independent datasets derived from the work of [12] were taken to develop and validate the proposed model PVPred-SCM for making a fair comparison with existing methods [8-16].In this study, the benchmark dataset consists of 99 PVPs (TR-positive.txt) and 208 non-PVPs (TR-negative.txt), while the independent dataset consists of 30 PVPs (Ind-positive.txt) and 64 non-PVPs (Ind-negative.txt). To validate the robustness of the proposed PVPred-SCM, we further tested and compared its performance using another independent dataset derived from Zhang et al’s method [11] (containing 68 PVPs and 92 non-PVPs) (Independent from Zhang et al..xlsx).