Loading...







Tools


  • GOntoSim

    • In this study, we propose GOntoSim, a novel method to determine the functional similarity between genes.
    • GOntoSim quantifies the similarity between pairs of GO terms, by taking the graph structure and the information content (IC) of nodes into consideration.
    • We achieved a purity score of 0.94 in contrast to 0.48 for the state-ofthe-art measures on the enzyme dataset.
    • Click here to access the web server
    • GOntoSim can be downloaded from Github
    • Reference:
  • Enzyme Function Prediction

    • HECNet: a hierarchical approach to Enzyme Function Classification using a Siamese Triplet Network

      • In this study, we propose a novel computational approach to predict an enzyme’s function up to the fourth level of the Enzyme Commission (EC) Number.
      • Our method uses innovative deep learning approaches along with an efficient hierarchical classification scheme to predict an enzyme’s precise function.
      • On a dataset of 11,353 enzymes and 402 classes, we achieved a hierarchical accuracy and Macro-F1 score of 91.2% and 81.9%, respectively, on the 4th level.
      • This methodology is broadly applicable for genome-wide prediction that can subsequently lead to automated annotation of enzyme databases and the identification of better/cheaper enzymes for commercial activities.
      • Click here to access the web server
      • Reference: Safyan Aman Memon, Kinaan Aamir Khan, Hammad Naveed. HECNet: a hierarchical approach to enzyme function classification using a Siamese Triplet Network, https://doi.org/10.1093/bioinformatics/btaa536
    • A hierarchical deep learning approach for multi-functional enzyme classification

      • In this study, we propose an approach to predict a multi-function enzyme’s function up to the most specific fourth level of the Enzyme Commission (EC) Number.
      • Our method uses a deep learning approach titled mlHECNet.
      • On a dataset of 2,583 multi-functional enzymes, we achieved a hierarchical subset accuracy and Macro-F1 score of 71.4% and 96.1%, respectively, on the 4th level.
      • Our method is broadly applicable and may be used to discover better enzymes.
      • Click here to access the web server
      • Reference: Kinaan Aamir Khan, Safyan Aman Memon, Hammad Naveed. A hierarchical deep learning approach for multi-functional enzyme classification, https://onlinelibrary.wiley/doi/full/10.1002/pro.4146
  • An integrated structure- and system-based framework to identify new targets of metabolites and known drugs.

    • A novel integrated structure- and system-based approach of drug-target prediction (iDTP) to enable the large-scale discovery of new targets for small molecules, such as pharmaceutical drugs, co-factors and metabolites (collectively called 'drugs').
    • In the cross-validation study, iDTP is used to predict the known targets of 11 drugs, with 63% sensitivity and 81% specificity.
    • This method is broadly applicable for the prediction of protein-small molecule interactions with several novel applications to biological research and drug development.
    • Download Here

    • Reference: Naveed H, Hameed US, Harrus D, Bourguet W, Arold ST, Gao X. An integrated structure- and system-based framework to identify new targets of metabolites and known drugs. Bioinformatics. 2015;31(24):3922-3929. doi: 10.1093/bioinformatics/btv477.
  • Predicting the pathogenicity of protein coding mutations using Natural Language Processing

    • Ths method uses NLP and text mining techniques to extract information embedded in PubMed articles to predict the functional imapact of coding mutations.
    • The performance of the methodology is tested on the manually curated OncoKB, VariBench datasets along with the two large automatically annotated datasets (CosmicVariSNP and Uniprot_humsavar) to check for generalizability.
    • The perfomance of the methodology is compared with the state of the art function prediction methods like polyphen2, mutation assessor and mutpred2. Our methodology outperforms these methods by just using NLP features.
    • This method will assist the oncologist in classifying the genetic mutation in less time and reduce toxicity in cancer patients.
    • Click here to access the web server

    • Reference: N. Rehmat, H. Farooq, S. Kumar, S. ul Hussain and H. Naveed, "Predicting the pathogenicity of protein coding mutations using Natural Language Processing," 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), Montreal, QC, Canada, 2020, pp. 5842-5846, doi: 10.1109/EMBC44109.2020.9175781.