New tools combine biological knowledge with machine learning to help researchers extract meaningful insights from complex OMICS data
summary
- Corneto is a new computational tool that helps researchers combine different types of biological data with previous biological knowledge to map how molecules, such as genes and proteins, interact within a cell.
- Corneto analyzes various samples together at once to show which biological processes are common and unique across cell types and conditions.
- Researchers use cornetto to identify signaling pathways associated with chemotherapy resistance in patients with ovarian cancer, to identify shared and cell-specific pathways in disease research.
Scientists and collaborators at EMBL-EBI at Heidelberg University have developed Corneto, a new computational tool that uses machine learning to gain meaningful insights from complex biological data. Corneto allows users to extract molecular networks – maps of interactions between genes, proteins and signaling pathways – by combining experimental data for various samples and conditions with previous biological knowledge such as signaling and metabolic networks. This helps us to better understand the mechanisms that lead cells to become healthy or sick.
Understanding how molecules interact within cells is key to uncovering mechanisms that may be wrong and potentially leading to disease. However, as the types of OMICS data available to researchers grow in size and complexity, researchers often struggle to extract useful and meaningful patterns from them. Corneto, short for constrained optimization for network recovery from OMICS, combines machine learning techniques with biological prior knowledge to simultaneously analyze multiple types of OMICS data, including trunk liptomic, proteomics, and metabolomics.
What does Omics mean?
OMICS refers to the large-scale study of biological molecules and their functions that use high-throughput technology to analyze complex datasets and their functions, as well as the functions within living systems. This includes areas such as genomics, transcriptomics, proteomics, metabolomics and more.
“We wanted to solve the general challenges of systems biology. How to understand OMICS data when very complex data is available at once,” said Julio Saez-Rodriguez, Head of Research at EMBL-EBI at Heidelberg University. “Corneto combines these complex data with previous information from biological databases to help find consistent, interpretable, and biologically meaningful patterns.”
Unified OMICS Analysis
Traditionally, scientists have compared healthy cells with diseased cells, for example, to analyze data from one condition at a time, building each individual interaction network. However, this approach can miss a bigger picture. Corneto uses machine learning to analyze multiple samples or conditions together, highlighting the biological processes shared between datasets and identifying differences between samples. Corneto is also designed to allow researchers to customize for specific use cases and extend to new data types if needed.
“Using a Cornet is like finding a common thread on the tangled web,” explained Pablo Rodriguez Meyer, a postdoctoral researcher at the University of Heidelberg. “It helps researchers to elicit the major biological processes that are happening in many samples and understand what is different or the same in each.”
Real World Applications
Using Corneto is especially valuable for researchers working in fields like cancer research where there are similarities between patients, but the two patients are not quite similar. To demonstrate this, researchers used Corneto to analyze gene expression data from multiple cancer patients and discovered which specific intracellular signaling pathways were operating abnormally.
Using transcriptomic data only, Corneto identified key deregulation kinases, enzymes that regulate cell signaling independently detected using phosphorus products. The resulting network revealed both shared pathways and patient-specific differences. This is a step towards the kind of insight that can one day support personalized treatment strategies.
Corneto is currently being used by EU research project decisioners to identify deregulated signaling pathways associated with chemotherapy resistance in patients with ovarian cancer.
Researchers also used Corneto to analyze metabolic pathways in yeast strains where different genes were inactivated. Here, Corneto was able to find the important process that yeast cells were using to survive and grow. Understanding these important processes can help scientists design better yeast strains and make biofuels and other products for industrial manufacturing.
Open source and ready for use
Corneto is available as open source software on GitHub. Here you can also find tutorials, examples of datasets, modular code and tailor the cornet to your needs.
Funds
This work was funded by the European Union's Horizon 2020 program under grants No. 951773 (Permedcoe) and No. 965193 (Decidinger).
/Public release. This material of the Organization of Origin/Author is a point-in-time nature and may be edited for clarity, style and length. Mirage.news does not take any institutional position or aspect, and all views, positions and conclusions expressed here are the views of the authors alone.
