SAPH-ire TFx - A Recommendation-based Machine Learning Model Captures a Broad Feature Landscape Underlying Functional Post-Translational Modifications

Abstract

Protein post-translational modifications (PTMs) are a rapidly expanding feature class of significant importance in cell biology. Due to a high burden of experimental proof, the number of functional PTMs in the eukaryotic proteome is currently underestimated. Furthermore, not all PTMs are functionally equivalent. Therefore, computational approaches that can confidently recommend the functional potential of experimental PTMs are essential. To address this challenge, we developed SAPH-ire TFx (https://saphire.biosci.gatech.edu/): a multi-feature neural network model and web resource optimized for recommending experimental PTMs with high potential for biological impact. The model is rigorously benchmarked against independent datasets and alternative models, exhibiting unmatched performance in the recall of known functional PTM sites and the recommendation of PTMs that were later confirmed experimentally. An analysis of feature contributions to model outcome provides further insight on the need for multiple rather than single features to capture the breadth of functional data in the public domain. Contactmtorres35@gatech.edu Supplementary InformationSee Tables S1-S6 & Figures S1-S4. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=189 SRC="FIGDIR/small/731026v2_figS1.gif" ALT="Figure 1"> View larger version (28K): org.highwire.dtl.DTLVardef@4dbe8dorg.highwire.dtl.DTLVardef@1936d5org.highwire.dtl.DTLVardef@15eef8aorg.highwire.dtl.DTLVardef@1c07cea_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOFigure S1.C_FLOATNO ROC curves at different KFSC thresholds. (unfurled in Figure 1E). C_FIG O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=121 SRC="FIGDIR/small/731026v2_figS4.gif" ALT="Figure 4"> View larger version (20K): org.highwire.dtl.DTLVardef@48e892org.highwire.dtl.DTLVardef@9cf372org.highwire.dtl.DTLVardef@32da4dorg.highwire.dtl.DTLVardef@99fd89_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOFigure S4.C_FLOATNO Model recall and recommendation comparison for PTMs associated with validated functional SLiMs from the ELM resource. C_FIG