{"entity": "researcher", "timestamp": "2026-08-20T20:58:15.528Z", "family": "Dwivedi", "given": "Sanjiv K", "initials": "SK", "orcid": "0000-0003-3400-4133", "affiliations": ["Bioinformatics, Department of Physics, Chemistry and Biology, Link\u00f6ping University, Link\u00f6ping, Sweden."], "links": {"self": {"href": "https://publications-affiliated.scilifelab.se/researcher/fa4b2a4afa3942beac5115b5ab505825.json"}, "display": {"href": "https://publications-affiliated.scilifelab.se/researcher/fa4b2a4afa3942beac5115b5ab505825"}}, "publications": [{"entity": "publication", "iuid": "7c1cb7d6bb354c10bd7b51364f4bb4d1", "links": {"self": {"href": "https://publications-affiliated.scilifelab.se/publication/7c1cb7d6bb354c10bd7b51364f4bb4d1.json"}, "display": {"href": "https://publications-affiliated.scilifelab.se/publication/7c1cb7d6bb354c10bd7b51364f4bb4d1"}}, "title": "Deriving disease modules from the compressed transcriptional space embedded in a deep autoencoder.", "authors": [{"family": "Dwivedi", "given": "Sanjiv K", "initials": "SK", "orcid": "0000-0003-3400-4133", "researcher": {"href": "https://publications-affiliated.scilifelab.se/researcher/fa4b2a4afa3942beac5115b5ab505825.json"}}, {"family": "Tj\u00e4rnberg", "given": "Andreas", "initials": "A", "orcid": "0000-0003-0064-1791", "researcher": {"href": "https://publications-affiliated.scilifelab.se/researcher/2d4d51b0b08948cd8be856ec4c0a50b2.json"}}, {"family": "Tegn\u00e9r", "given": "Jesper", "initials": "J", "orcid": "0000-0002-9568-5588", "researcher": {"href": "https://publications-affiliated.scilifelab.se/researcher/b115e5faf68a4acbb6f947c711fd37ec.json"}}, {"family": "Gustafsson", "given": "Mika", "initials": "M", "orcid": "0000-0002-0048-4063", "researcher": {"href": "https://publications-affiliated.scilifelab.se/researcher/d87fbeece2db4a9dbaf2254bfd9a1100.json"}}], "type": "journal article", "published": "2020-02-12", "journal": {"title": "Nat Commun", "issn": "2041-1723", "volume": "11", "issue": "1", "pages": "856", "issn-l": "2041-1723"}, "abstract": "Disease modules in molecular interaction maps have been useful for characterizing diseases. Yet biological networks, that commonly define such modules are incomplete and biased toward some well-studied disease genes. Here we ask whether disease-relevant modules of genes can be discovered without prior knowledge of a biological network, instead training a deep autoencoder from large transcriptional data. We hypothesize that modules could be discovered within the autoencoder representations. We find a statistically significant enrichment of genome-wide association studies (GWAS) relevant genes in the last layer, and to a successively lesser degree in the middle and first layers respectively. In contrast, we find an opposite gradient where a modular protein-protein interaction signal is strongest in the first layer, but then vanishing smoothly deeper in the network. We conclude that a data-driven discovery approach is sufficient to discover groups of disease-related genes.", "doi": "10.1038/s41467-020-14666-6", "pmid": "32051402", "labels": [], "xrefs": [{"db": "pmc", "key": "PMC7016183"}, {"db": "pii", "key": "10.1038/s41467-020-14666-6"}], "notes": [], "created": "2026-08-20T08:51:03.573Z", "modified": "2026-08-20T08:51:03.799Z"}]}