March 28, 2024

The Human Cell Atlas: AI Helps To Spot Single Diseased Cells

Mapping brand-new mates of cells of healthy people and COVID-19 patients onto a healthy cells reference atlas. Credit: Helmholtz Zentrum München/ Mohammad Lotfollahi
The Human Cell Atlas is the worlds largest, growing single-cell referral atlas. These references assist physicians to comprehend the influences of aging, environment and disease on a cell– and eventually detect and deal with patients much better.
Scientists from Helmholtz Zentrum München and the Technical University of Munich (TUM) established a novel algorithm called “scArches,” short for single-cell architecture surgery. The most significant benefit: “Instead of sharing raw information between clinics or proving ground, the algorithm uses transfer finding out to compare new datasets from single-cell genomics with existing referrals and thus maintains privacy and privacy. This also makes annotating and analyzing of brand-new data sets really easy and equalizes the usage of single-cell recommendation atlases significantly,” states Mohammad Lotfollahi, the leading researcher of the algorithm.
Example COVID-19
The scientists used scArches to study COVID-19 in several lung bronchial samples. They compared the cells of COVID-19 patients to healthy recommendations using single-cell transcriptomics. The algorithm was able to different unhealthy cells from the references and thus allowed the user to identify the cells in requirement of treatment, for both mild and extreme COVID-19 cases. Biological variation between patients did not impact the quality of the mapping process.

Fabian Theis: “Our vision is that in the future we will utilize cell recommendations as easily as we nowadays provide for genome references. In another word, if you want to bake a cake, you generally do not wish to attempt creating your own recipe– rather you just look one up in a cookbook. With scArches, we formalize and simplify this lookup process.”
Discover more about scArches: https://github.com/theislab/scarches
Reference: “Mapping single-cell information to reference atlases by transfer knowing” by Mohammad Lotfollahi, Mohsen Naghipourfar, Malte D. Luecken, Matin Khajavi, Maren Büttner, Marco Wagenstetter, Žiga Avsec, Adam Gayoso, Nir Yosef, Marta Interlandi, Sergei Rybakov, Alexander V. Misharin and Fabian J. Theis, 30 August 2021, Nature Biotechnology.DOI: 10.1038/ s41587-021-01001-7.
Computational biologist Mohammad Lotfollahi is a group leader at Fabian Theis lab at Helmholtz Zentrum München and doctoral student at TUM School of Life Sciences at the Technical University of Munich. He works closely with Fabian Theis, who is Director of the Institute of Computational Biology at Helmholtz Zentrum München and Coordinator of the Helmholtz Artificial Intelligence Cooperation Unit (Helmholtz AI). Theis holds the chair for Mathematical Modelling of Biological Systems at TUM.

The Human Cell Atlas is the worlds largest, growing single-cell recommendation atlas. They compared the cells of COVID-19 patients to healthy recommendations utilizing single-cell transcriptomics. The algorithm was able to different diseased cells from the references and therefore enabled the user to identify the cells in requirement of treatment, for both moderate and severe COVID-19 cases. Fabian Theis: “Our vision is that in the future we will use cell references as quickly as we nowadays do for genome referrals.