Latent space arithmetic on data embeddings from healthy multi-tissue human RNA-seq decodes disease modulesShow others and affiliations
2024 (English)In: Patterns, ISSN 2666-3899, Vol. 5, no 11, article id 101093Article in journal (Refereed) Published
Abstract [en]
The human transcriptome is a highly complex system and is often the focus of research, especially when it fails to function properly, causing disease. Indeed, the amount of publicly available transcriptomic data has grown considerably with the advent of high-throughput techniques. Such special cases are often hard to fully dissect, since studies will be confined to limited data samples and multiple biases. An ideal approach would utilize all available data to learn the fundamentals of the human gene expression system and use these insights in the examination of the more limited sample sets relating to specific diseases. This study shows how a neural network model can be created and used to extract relevant disease genes when applied to limited disease datasets and to suggest relevant pharmaceutical compounds. Thus, it presents a step toward a future where artificial intelligence can advance the analysis of human high-throughput data.
Place, publisher, year, edition, pages
Elsevier, 2024. Vol. 5, no 11, article id 101093
National Category
Bioinformatics and Computational Biology Bioinformatics (Computational Biology) Medical Genetics and Genomics
Research subject
Bioinformatics
Identifiers
URN: urn:nbn:se:his:diva-24648DOI: 10.1016/j.patter.2024.101093ISI: 001355226900001PubMedID: 39568475Scopus ID: 2-s2.0-85208221759OAI: oai:DiVA.org:his-24648DiVA, id: diva2:1910326
Funder
Hedlund foundation, M-2023-2054University of SkövdeSwedish Research Council, 2019-04193Swedish Research Council, 2022-06725Stiftelsen Assar Gabrielssons fond, FB21-66Knowledge Foundation, 20200014
Note
CC BY 4.0
Available online 31 October 2024, 101093
Correspondence: hendrik.de.weerd@liu.se (H.A.d.W.), rasmus.magnusson@liu.se (R.M.)
This work was supported by the Systems Biology Research Centre at the University of Skövde under grants from the Swedish Knowledge Foundation (grant 20200014 to R.M, Z.L.-P., and J.S.), Petrus och Augusta Hedlunds Stiftelse (grant M-2023-2054 to R.M), the Assar Gabrielssons Fond (grant FB21-66 to R.M. and H.A.d.W.), and the Swedish Research Council (grant 2019-04193 to H.A.d.W. and M.G.). The computations were enabled by resources provided by the Swedish National Infrastructure for Computing (SNIC), partially funded by the Swedish Research Council through grant agreement no. 2022-06725.
2024-11-042024-11-042025-02-10Bibliographically approved