AI Analytics Find New ALS Genetic Drug Targets

(Gerd Altmann, Pixabay)

8 July 2022. Researchers analyzed medical and scientific databases with algorithms to identify eight previously unreported biomarkers contributing to neurodegeneration similar to ALS. A team from Insilico Medicine in New York and academic labs in China, Europe, and the U.S. reported their findings last week in the journal Frontiers in Aging Neuroscience.

Amyotrophic lateral sclerosis or ALS is a progressive neurological disorder that attacks neurons or nerve cells controlling voluntary muscle movements throughout the body. People with ALS, also known as Lou Gehrig’s disease, experience weakened muscles as neurons fail to receive signals from the brain, leading to loss of their ability to use arms, legs, and facial muscles. As loss of muscular control extends to the lungs, people with ALS lose their ability to breathe without external support. Most deaths from ALS are from respiratory failure.

Insilico Medicine offers analytics services with deep learning and other artificial intelligence tools for drug discovery and development. One focus of the company is the aging process, including effects of neurodegeneration. As reported by Science & Enterprise in Nov. 2019, Insilico Medicine takes part in a consortium on applying A.I. to study and extend human longevity. One of the services provided by the company, called PandaOmics, uses deep learning algorithms to identify new drug targets, from analyzing genomic, protein, and clinical data sets.

Targets screened in preclinical model organisms

In their paper, the team applied PandaOmics to analyze data from a number of databases with genetic, transcription, protein findings, as well as samples of motor neurons derived from stem cells taken post-mortem from 135 ALS patients and 31 other people for comparison. The motor neuron samples were provided by Answer ALS, an organization promoting advanced integrated research on ALS therapies, including transformation of adult stem cells into motor neurons as tools for better understanding neurodegeneration.

The researchers performed separate deep-learning analyses of data from 237 ALS patients and 91 other individuals for comparison, as well as sub-samples of patients divided by sporadic-onset or familial ALS. After a series of analytic runs, PandaOmics identified 28 potential biomarker targets. The team then screened the target candidates in drosophila, a common preclinical model organism, with edited C9ORF72 genes to mimic common mutations associated with ALS, which validated 18 of those targets.

The 18 targets identified and validated by the team include a mix of known genetic conditions associated with ALS as well as new targets not previously linked to the disease. The screening with edited C9ORF72 genes in drosophila revealed eight genes not associated earlier with ALS. Further tests with the model species show suppression of those genes stops degeneration of neurons in drosophila eye tissue.

Insilico Medicine plans to advance the identified targets for further development as ALS treatment candidates, and post the data on its ALS.ai web site. “We are working with collaborators to progress some targets toward clinical trials for ALS,” says Insilico co-CEO and chief scientist Feng Ren in a company statement released through Globe Newswire. “At the same time, we are also further expanding the utilization of PandaOmics to discover novel targets for other disease areas including oncology, immunology, and fibrosis.”

More from Science & Enterprise:

• FDA Issues Action Plan for Rare Neuro Diseases
• ALS Therapy Start-Up Gains $5.1M in Seed Funds
• Preclinical Test Advances Synthetic Bio ALS Treatment
• Engineered Virus Delivers Brain-Targeted Gene Therapy
• Oxford, GSK Partner on New Tech for Neuro Diseases

We designed Science & Enterprise for busy readers including investors, researchers, entrepreneurs, and students. Except for a narrow cookies and privacy strip for first-time visitors, we have no pop-ups blocking the entire page, nor distracting animated GIF graphics. If you want to subscribe for daily email alerts, you can do that here, or find the link in the upper left-hand corner of the desktop page. The site is free, with no paywall. But, of course, donations are gratefully accepted.

*     *     *