AI and multiomics identify cancer biomarker and drug target

AI and multiomics identify cancer biomarker and drug target


Source: MasterTux/Pixabay

Artificial intelligence (AI) machine learning has the potential to completely revolutionize not only how diseases are detected, but also how to identify personalized treatment based on a patient’s genomics. A new study published in Cell death and diseasepeer review Nature journal, shows how an AI-based multiomics platform can identify novel biomarkers for new therapeutic targets from gene expression signatures in cancer-associated diseases.

“The identification of novel molecular biomarkers stratifying cancer patients with different survival outcomes may provide new opportunities for target discovery and the subsequent development of tailored therapies,” wrote the researchers from Insilico Medicine, University of Copenhagen and the University of Chicago.

To find these biomarkers, the researchers used Insilico Medicine’s AI-based platform called PandaOmics to find novel cancer targets and analyze gene expression mutations in rare diseases deficient in cell repair. DNA. PandaOmics is an AI deep learning algorithm that reads scientific papers to produce a graphical representation of the content. It also allows comparison between data sets, harmonization of data and analysis of activation or inhibition pathways.

The AI ​​algorithm enables drug discovery by identifying therapeutic targets for diseases by analyzing relevant data and genes. It draws on five million samples of omics data (transcriptomics, genomics, epigenomics, proteomics, single-cell data), in addition to relevant data from over 3.8 million patents, 30 million published biomedical research, 1, 3 million drugs from early phase clinical trial to launch phase, 342,000 clinical trials and three million life science research grants funded.

In science, omics refers to fields of study that end in -omics, such as genomics (study of an organism’s genomes), neurogenetics (study of the genetic impact on the nervous system), psychogenomics ( applied genomics and proteomics to understand the impact on normal and diseased brain and behavior), microbiomics (study of the genomes of microorganisms) and connectomics (study of neuronal connections in the brain – the connectome).

Other examples of omics include pangenomics (study of all genes in a species), lipdomics (study of lipid pathways and networks), immunoproteomics (study of proteins and immune response), glycomics ( study of sugars and carbohydrates), pharmacogenomics (study of the genome and the response to drugs), toxicogenomics (study of the activity of genes and proteins in cells or tissues exposed to toxins), metabolomics (study of chemical processes of metabolites), transcriptomics (the study of RNA transcriptions of an organism – transcriptome), proteomics (study of proteins), epigenomics (study of all the epigenetic modifications on the genetic material of a cell-epigenome), etc.

Multiomics is the integration of a variety of omics into a single analysis. The integration of artificial intelligence machine learning with multiomics data analysis has enabled scientists to rapidly discover new biomarkers.

“To select diseases for further gene expression analysis and identification of novel cancer biomarkers, we performed hierarchical clustering based on analysis of common clinical phenotypes prevalent in several cell repair diseases. ‘DNA,” the researchers wrote. “Notably, we found three main clusters of diseases spanning various phenotypes.”

The scientists identified three main disease groups and selected these rare inherited diseases for further analysis: Louis-Bar syndrome (ataxia-telangiectasia), Nijmegen rupture syndrome and Werner syndrome. All three are autosomal recessive diseases, meaning two copies of the abnormal gene must be present for the traits or syndrome to develop, according to the US National Institute of Health (NIH).

Louis-Bar syndrome is a rare inherited neurodegenerative disease that causes severe disability and affects body systems such as the nervous system and the immune system. People with Louis-Bar syndrome are at high risk for developing cancers of the immune system and certain types of blood cancers, according to St. Jude Children’s Research Hospital.

Nijmegen Breakage Syndrome is a rare genetic disorder that presents at birth with unusually small head size (microcephaly), dysmorphic facial features, and short stature. The name comes from the multitude of DNA breaks found in people with the disease. According to the NIH, people with Nijmegen’s rupture syndrome have an increased risk of cancer, particularly of developing non-Hodgkin’s lymphoma, cancer of cells of the immune system, and other cancers associated with this syndrome, including cancers brain such as glioma and medulloblastoma, as well as rhabdomyosarcoma, a cancer of muscle tissue.

Werner syndrome is a rare disease characterized by premature aging and an increased risk of cancers such as skin and thyroid cancers. The most common causes of death for people with Werner syndrome are cancer and heart attacks, according to the NIH.

The researchers used gene expression datasets from DNA repair deficient disorders with high cancer risk to identify biomarkers for frequently dysregulated genes that may be associated with cancer progression. In order to spot cancer-related pathways in the diseases, the team analyzed changes in gene expression profiles, particularly those that showed dysregulated genes.

“Notably, CEP135 was the most down-regulated gene with a similar expression pattern in all three DNA repair diseases, suggesting that it may be associated with the common cancer phenotype,” the authors reported. scientists.

The scientists hypothesized that the CEP135 gene could serve as a predictive biomarker that could classify patients into subgroups with different survival outcomes. To test this, the team performed a survival analysis for 33 types of cancer from The Cancer Genome Atlas (TCGA) dataset, one of the largest and most comprehensive genomic datasets available. with cancer samples from more than 11,000 patients spanning a twelve-year period according to the United States. National Cancer Institute.

This led to the discovery that the CEP135 gene can serve as a predictive biomarker for patients with sarcoma, a rare type of cancerous tumor that grows in connective tissue. Refining their analysis on sarcoma patients with high CEP135 gene expression and low survival outcomes, the scientists using the AI ​​algorithm discovered a list of the 20 most promising target genes. This list of candidate genes was then reduced to five genes after experimental verification. Of these five others, the polo-like kinase 1 (PLK1) gene was the only one to show a significant decrease in cell growth, making it a potential target for future cancer therapy.

“Although further target validation is needed, this study demonstrated the potential of in silico studies for rapid biomarker discovery and target characterization,” the scientists concluded.

Copyright © 2022 Cami Rosso All rights reserved.

#multiomics #identify #cancer #biomarker #drug #target

Leave a Comment

Your email address will not be published. Required fields are marked *