Optimizing InterProScan feature processing generates a surprisingly good protein function prediction method
Tiittanen, H.; Holm, L.; Toronen, P.
Show abstract
MotivationAutomated protein Function Prediction (AFP) is an intensively studied topic. Most of this research focuses on methods that combine multiple data sources, while fewer articles look for the most efficient ways to use a single data source. Therefore, we wanted to test how different preprocessing methods and classifiers would perform in the AFP task when we process the output from the InterProscan (IPS). Especially, we present novel preprocessing methods, less used classifiers and inclusion of species taxonomy. We also test classifier stacking for combining tested classifier results. Methods are tested with in-house data and CAFA3 competition evaluation data. ResultsWe show that including IPS localisation and taxonomy to the data improves results. Also the stacking improves the performance. Surprisingly, our best performing methods outperformed all international CAFA3 competition participants in most tests. Altogether, the results show how preprocessing and classifier combinations are beneficial in the AFP task. Contactpetri.toronen(AT)helsinki.fi Supplementary informationSupplementary text is available at the project web site http://ekhidna2.biocenter.helsinki.fi/AFP/ and at the end of this document.
Matching journals
The top 6 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- A cautionary tale about properly vetting datasets used in supervised learning predicting metabolic pathway involvement 96%
- BioDiscViz : a visualization support and consensus signature selector for BioDiscML results 95%
- Automated recognition of functional compound-protein relationships in literature 95%
Similar papers in this journal
- Mining drug-target interactions from biomedical literature using chemical and gene descriptions-based ensemble transformer model. 96%
- Batch-effect correction in single-cell RNA sequencing data using JIVE 95%
- Improving protein function prediction by learning and integrating representations of protein sequences and function labels 95%
Similar papers in this journal
- DeepInsight-3D for precision oncology: an improved anti-cancer drug response prediction from high-dimensional multi-omics data with convolutional neural networks 95%
- Finding disease modules for cancer and COVID-19 in gene co-expression networks with the Core&Peel method 94%
- Tensor decomposition- and principal component analysis-based unsupervised feature extraction to select more reasonable differentially expressed genes: Optimization of standard deviation versus state-of-art methods 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.