Back

Comprehensive multi-omics profiling of a healthy human cohort

Visser, C.; Niehues, A.; Najdekr, L.; Toth, R.; Nazarov, P. V.; Vrbkova, J.; Stankova, J.; Ekberg, S.; Moreno, E. C.; Ghimire, B.; Yadav, B.; Mattila, P.; Puhka, M.; Lanza, V. F.; Gloerich, J.; Kulkarni, P.; Wessels, H.; Engelke, U.; Oldoni, E.; Andreu, T.; Saunders, G.; Muller, A.; Bendova, M.; Rozankova, Z.; Dzubak, P.; Pavlis, P.; Siska, B.; Spantidaki, A.; Beisly, D.; Srovnal, J.; Moravikova, V.; Sestak, P.; Gazdarica, M.; Scherer, A.; Bermejo, M. L. G.; Nordlund, J.; Gool, A.; Hajduch, M.; 't Hoen, P. A. C.

2025-12-09 bioinformatics
10.1101/2024.11.07.622407 bioRxiv
Show abstract

Multi-omics approaches can offer powerful insights into personalized biomarker profiles relevant for disease diagnosis, prognosis, and therapeutics. However, separating meaningful biological variability from technical noise remains a major challenge. The EATRIS-Plus consortium analyzed blood samples from 127 healthy adults across six omics layers using twelve platforms, resulting in one of the most comprehensive multi-omics profiling datasets of healthy individuals available to date. We applied reproducible workflows to analyze and integrate these data, revealing several key findings. Sex significantly influenced all omics layers, emphasizing the importance of sex-balanced study designs. Age could be accurately predicted using epigenetic clocks, achieving high performance with our high-resolution enzymatic methylation sequencing data (R2 = 0.90), whereas candidate aging biomarkers were identified across all omics layers. The resulting dataset provides reference ranges in healthy individuals for abundance and variability of omics features, enabling robust power analyses, sample size estimations, and benchmarking of multi-omics integration methods. This resource can guide future biomarker discovery and personalized health research and was made FAIR-compliant and publicly available via the ClinData Portal (https://clindata.imtm.cz) and a Zenodo repository (https://doi.org/10.5281/zenodo.17514796).

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"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.