42 Post-translational Modifications (PTMs), chemical or proteinaceous covalent alterations to 43 the side chains of amino acid residues in proteins, are a rapidly expanding feature class of 44 significant importance in cell biology. Due to a high burden of experimental proof and the lack of 45 effective means for experimentalists to prioritize PTMs by functional significance, currently less 46 than ~2% of all PTMs have an assigned biological function. Here, we describe a new artificial 47 neural network model, SAPH-ire TFx for the functional prediction of experimentally observed 48 eukaryotic PTMs. Unlike previous functional PTM prioritization models, SAPH-ire TFx is optimized 49 with both receiver operating characteristic (ROC) and recall metrics that maximally capture the 50 range of diverse feature sets comprising the functional modified eukaryotic proteome. The tool 51 was through systematic evaluation of input features, model architectures, training procedures, 52 and interpretation metrics using a 2018 training dataset of 430,750 PTMs containing 7,480 PTMs 53 with literature-supported evidence of biological function. The resulting model was used to classify 54 an expanded 2019 dataset of 512,015 PTMs (12,867 known functional) containing 102,475 PTMs 55 unencountered in the original dataset. Model output from the 2019 extended dataset was 56 benchmarked against pre-existing prediction models, revealing superior performance in 57 classification of functional and/or disease-linked PTM sites. Finally, a dynamic web interface 58 provides customizable graphical and tabular visualization of PTM and SAPH-ire TFx data within 59 the context of all modifications within a protein family, exposing several metrics by which important 60 functional PTMs can be identified for investigation. 61 62 63 64 AUTHOR SUMMARY 65The modification of proteins after they are translated is an important process that can 66 control the structure and function of the proteins on which they occur. Hundreds of different types 67 of modification happen at some point during the lifetime of every protein in eukaryotic cells and 68play an essential role in cellular processes such as cell division, cell communication, gene 69regulation. Using current state-of-the-art detection tools, the rate at which post-translational 70 modifications are detected now far surpasses the rate at which they can be investigated for 71functionality. Furthermore, not all modifications detected are functional, making it difficult to 72determine into which modifications one should invest experimental effort. Here, we describe a 73 new computational tool -SAPH-ire TFx -capable of predicting functional modification sites from 74 large-scale datasets, and consequently focus experimental effort towards only those 75 modifications that are likely to be biologically significant. We show that the tool performs well 76 across multiple datasets within which known functional modifications are scattered; and we show 77 that the tool outperforms prior functional prioritization tools...