Algorithms help to differentiate diseases at the molecular level

Machine learning is playing an increasing role in biomedical research. Scientists from the Technical University of Munich (TUM) have developed a new method for using molecular data to extract subtypes of diseases. In the future, this method may help to support the study of larger groups of patients.

Nowadays, doctors identify and diagnose most diseases based on symptoms. However, this does not necessarily mean that the diseases of patients with similar symptoms will have identical causes or will demonstrate the same molecular changes. Molecular mechanisms of the disease are often discussed in biomedicine. This refers to changes in the regulation of genes, proteins or metabolic pathways at the onset of the disease. The goal of stratified medicine is to classify patients into different subtypes at the molecular level to provide more targeted treatment.

To extract disease subtypes from large patient datasets, new machine learning algorithms may help. They are designed to independently recognize patterns and correlations in extensive clinical measurements. The junior research group LipiTUM, led by Dr. Josh Constantine Pauling of the Department of Experimental Bioinformatics, has developed an algorithm for this purpose.

Complex analysis using an automated web tool

Their method combines the results of existing algorithms to obtain more accurate and stable predictions for clinical subtypes. This combines the features and benefits of each algorithm and eliminates their time-consuming correction. “This makes it much easier to apply the analysis to clinical trials,” said Dr. Pauling. “For this reason, we have developed a web-based tool that allows online analysis of molecular clinical data by practitioners without prior knowledge of bioinformatics.

On the website (https://exbio.wzw.tum.de/mosbi/) researchers can send their data for automated analysis and use the results to interpret their research. “Another important aspect for us was the visualization of the results. Previous approaches have not been able to generate intuitive visualizations of patient relationships, clinical factors, and molecular signatures. That will change with the web-based visualization produced by our MoSBi tool, ”said Tim Rose, a scientist at TUM School of Life Sciences. MoSBi means “Molecular signatures using biclastering”. “Biclustering” is the name of the technology used by the algorithm.

Application for clinically relevant issues

With the tool, researchers can now, for example, present data from cancer studies and simulations for different scenarios. They have already demonstrated the potential of their method in a large-scale clinical trial. In a joint study with researchers from the Max Planck Institute in Dresden, the Technical University of Dresden and the University Clinic in Kiel, they studied the change in lipid metabolism in the liver of patients with non-alcoholic fatty liver disease (NAFLD).

This widespread disease is associated with obesity and diabetes. It develops from non-alcoholic fatty liver (NAFL), in which lipids are deposited in the liver cells, to non-alcoholic steatohepatitis (NASH), in which the liver is further inflamed, to cirrhosis of the liver and tumor formation. Apart from dietary adjustments, no treatments have been found so far. Because the disease is characterized and diagnosed by the accumulation of various lipids in the liver, it is important to understand their molecular composition.

Biomarkers for liver disease

Using MoSBi methods, the researchers were able to demonstrate the heterogeneity of the liver of patients in the NAFL stage at the molecular level. “From a molecular point of view, the liver cells of many NAFL patients were almost identical to those of NASH patients, while others were still largely similar to healthy patients. We could also confirm our predictions using clinical data, “said Dr. Pauling. “We were then able to identify two potential lipid biomarkers for disease progression.” This is important for early detection of the disease and its progression and the development of targeted treatments.

The research group is already working on further applications of its method to gain a better understanding of other diseases. “In the future, algorithms will play an even bigger role in biomedical research than they do today. They can make it much easier to find complex mechanisms and find more targeted treatment approaches, ”says Dr. Pauling.

reference: Rose TD, Bechtler T, Ciora OA and others. MoSBi: Automatic digging of molecular stratification and subtyping signatures. Proc Natl Acad Sci. 2022; 119 (16): e2118210119. doi: 10.1073 / pnas.2118210119

This article has been republished as follows materials. Note: The material can be edited for length and content. For more information, please contact the cited source.

Leave a Comment

Your email address will not be published.