Web-based and machine learning approaches for identification of patient-reported outcomes in inflammatory bowel disease.

BACKGROUND: Messages from an Internet forum are raw material that emerges in a natural setting (i.e., non-induced by a research situation). AIMS: The FLARE-IBD project aimed at using an innovative approach consisting of collecting messages posted by patients in an Internet forum and conducting a machine-learning study (data analysis/language processing) for developing a patient-reported outcome measuring flare in inflammatory bowel disease meeting international requirements. METHODS: We used web-based and machine learning approaches, in the following steps. 1) Web-scraping to collect all available posts in an Internet forum (23 656 messages) and extracting metadata from the forum. 2) Twenty patients were randomly assigned 50 extracted messages; participants indicated whether the message corresponded or not to the flare phenomenon (labeling). If yes, participants were asked to identify excerpts from the text they considered significant flare markers (annotation). 3) The set of annotated messages underwent a vocabulary analysis. RESULTS: The phenomenon of flare was circumscribed with the identification of 20 surrogate flare markers classified into five dimensions with their frequency within extracted labeled data: impact on life, symptoms, extra-intestinal manifestations, drugs and environmental factors. Web-based and machine-learning approaches met international recommendations to inform the content and structure for the development of patient-reported outcomes.

PPIDomainMiner: Inferring domain-domain interactions from multiple sources of protein-protein interactions.

Many biological processes are mediated by protein-protein interactions (PPIs). Because protein domains are the building blocks of proteins, PPIs likely rely on domain-domain interactions (DDIs). Several attempts exist to infer DDIs from PPI networks but the produced datasets are heterogeneous and sometimes not accessible, while the PPI interactome data keeps growing. We describe a new computational approach called "PPIDM" (Protein-Protein Interactions Domain Miner) for inferring DDIs using multiple sources of PPIs. The approach is an extension of our previously described "CODAC" (Computational Discovery of Direct Associations using Common neighbors) method for inferring new edges in a tripartite graph. The PPIDM method has been applied to seven widely used PPI resources, using as "Gold-Standard" a set of DDIs extracted from 3D structural databases. Overall, PPIDM has produced a dataset of 84,552 non-redundant DDIs. Statistical significance (p-value) is calculated for each source of PPI and used to classify the PPIDM DDIs in Gold (9,175 DDIs), Silver (24,934 DDIs) and Bronze (50,443 DDIs) categories. Dataset comparison reveals that PPIDM has inferred from the 2017 releases of PPI sources about 46% of the DDIs present in the 2020 release of the 3did database, not counting the DDIs present in the Gold-Standard. The PPIDM dataset contains 10,229 DDIs that are consistent with more than 13,300 PPIs extracted from the IMEx database, and nearly 23,300 DDIs (27.5%) that are consistent with more than 214,000 human PPIs extracted from the STRING database. Examples of newly inferred DDIs covering more than 10 PPIs in the IMEx database are provided. Further exploitation of the PPIDM DDI reservoir includes the inventory of possible partners of a protein of interest and characterization of protein interactions at the domain level in combination with other methods. The result is publicly available at http://ppidm.loria.fr/.