Have a question? Call 952.469.8240   |   Need service or parts? Call 952.469.8230

AI Technology Can Identify Cancer Drug Targets

Facebook
Twitter
LinkedIn
cancer cells

According to a recent study published in Cell Systems, researchers from Weill Cornell Medicine have used artificial intelligence to identify drug targets based on setting maps of regulatory networks in patient tumors. Four drug candidates were identified and validated for liver, neuroendocrine, and renal cancers.

The associate professor of physiology and biophysics, Dr. Ekta Khurana, is a leader of the study. The team mapped gene regulatory networks for tumor samples from 371 patients with 22 cancer types by a new computational approach. Gene regulatory networks are models that help to describe the complex relationships between genes in a cell. Gene regulatory networks are often changed when someone gets cancer.

In order to build accurate gene regulatory networks, researchers incorporated data from the tumor cells on messenger RNA, which are transformed into proteins and chromatin accessibility, which can help reveal the way DNA packaging and other factors impact gene expression.

The team developed an innovative computational approach, called Cancer Regulatory Networks and Susceptibilities (CaRNetS), to find out key proteins that can be drug targets for cancer therapy within the gene regulatory networks. As a result, they established the identities of known targets like CTNNB1 (B-Catenin) in the colon, ERBB2 (Her2) in lung cancers, and BRAF in the skin.

Later, the scientists applied the computational approach aiming to find the key transcription factors and their interacting proteins, which may be points that can be targeted to stop or slow tumor growth. Transcription factors are proteins that connect to specific DNA sequences and adjust the expression of genes, turning their production on and off.

By using CaRNets on patient tumor samples, the researchers were able to cluster patients into 22 groups, including nine coped to only one cancer type and 13 contained patients from multiple cancer types.

The approach revealed drug targets for all 22 clusters allowing the research team to validate four of these protein candidates in cells. They found that inhibiting the proteins identified made a significant impact on growth of the cell lines, which represent renal, liver, and neuroendocrine cancer types when compared to controls.

With the ease of measuring chromatin accessibility from patient tissue on a large scale, the researchers believe that their computational approach will be widely used to find novel treatment options for more cancer types and subtypes.

Article Source: Technology Networks
Image by Mahmoud Ahmed from Pixabay