AI-Powered Novel Therapeutic Target Discovery

This approach has proven useful for a rare disease, but what about aging?


AI Drug DiscoveryAI Drug Discovery

A new study conducted by Insilico Medicine in collaboration with leading aging research institutions has identified novel therapeutic targets for amyotrophic lateral sclerosis, a devastating age-associated neurodegenerative disease [1].

To do this, we need your support. Your charitable contribution tranforms into rejuvenation research, news, shows, and more. Will you help?

Disease and aging: not that different

Amyotrophic lateral sclerosis (ALS) is a rare neuromuscular disease. It is most famous for afflicting the well-known physicist Stephen Hawking, and the 2014 ALS Ice Bucket Challenge raised millions of dollars for ALS research. The disease is characterized by motor neuron degeneration, which leads to progressive muscle atrophy, paralysis, and death, typically within 3-10 years from symptom onset. Importantly, aging is the most prevalent risk factor for developing ALS.

There are familial and sporadic cases of ALS. The former are associated with mutations in a number of genes, of which SOD1, TARDBP, C9orf72, and FUS are implicated in ~50% of all ALS cases. Overall, however, the genetic basis and the pathophysiological mechanisms of ALS are still far from being understood.

Nevertheless, similarly to so many other neurodegenerative diseases, both age-related and not, ALS is characterized by several processes that also underlie aging, such as loss of proteostasis, mitochondrial dysfunction, DNA repair deficiency, and neuroinflammation. Therefore, ALS therapeutic target identification might also be beneficial in addressing the hallmarks of aging.

In this study, the researchers applied the AI-driven discovery platform PandaOmics to identify novel potential therapeutic targets based on the altered molecular pathways in ALS.

Big data to answer big questions

The researchers utilized two types of datasets collected from ALS patients and healthy controls: post-mortem CNS tissues and motor neurons that were differentiated from induced pluripotent stem cells that had been generated from blood cells. In both types of datasets, the ALS patients were classified into either familial or sporadic subcategories, based on their genotype.

The proteomic and transcriptomic data retrieved from these datasets were then used to identify therapeutic targets by employing the PandaOmics platform. This cloud-based platform utilizes deep learning models and AI approaches to predict genes associated with diseases based on a combination of scores. Interestingly, the scores did not just include the raw molecular and -omics data, they included such parameters as grant sizes and the impact factors of the journals in which the relevant papers were published.

Various built-in PandaOmics filters were then applied to refine the target search. For example, druggability filters assess how safe and easy it would be to manipulate the identified target. By restricting certain filters and disabling certain scores, the researchers came up with 50 target genes that are dysregulated in ALS compared to controls.

Precious few

Further analysis yielded 28 potential candidates, with 17 high-confidence targets and 11 novel therapeutic targets. Among these, the researchers then verified 8 previously unreported genes as promising targets in vivo: the suppression of their homologs rescued eye neurodegeneration in a Drosophila model of ALS.

These genes were found to be upregulated in ALS, and their dysfunction reflects possible pathological mechanisms of the disease, including loss of proteostasis (ADRA2B, PPP3CB), inflammation (NR3C1, P2RY14, PTPRC), excitotoxicity (KCNB2, KCNS3), and impaired neural differentiation (RARA). Importantly, some of these genes, e.g. NR3C1, were also shown to be age-related.

In addition to individual genes, the researchers analyzed the dysregulated molecular pathways in ALS patients. Pathways relating to the innate immune system, programmed cell death, and the unfolded protein response were all activated in ALS tissue samples. Notably, the dysregulated pathways identified in post-mortem tissue samples differed from those identified in the motor neurons. The authors explain that the latter purely reflect the disease pathology without the aging context present in the former.

There were also differences in the dysregulated pathways between familial and sporadic ALS groups, with a higher intra-group variability in the latter. This is in line with the complex genetic basis of sporadic ALS, which probably explains the symptom variability in ALS individuals that is so typical for many neurological diseases. This calls for further investigation and will probably lead to personalized treatment options in the future.


Amyotrophic lateral sclerosis (ALS) is a severe neurodegenerative disease with ill-defined pathogenesis, calling for urgent developments of new therapeutic regimens. Herein, we applied PandaOmics, an AI-driven target discovery platform, to analyze the expression profiles of central nervous system (CNS) samples (237 cases; 91 controls) from public datasets, and direct iPSC-derived motor neurons (diMNs) (135 cases; 31 controls) from Answer ALS. Seventeen high-confidence and eleven novel therapeutic targets were identified and will be released onto ALS.AI (http://als.ai/). Among the proposed targets screened in the c9ALS Drosophila model, we verified 8 unreported genes (KCNB2, KCNS3, ADRA2B, NR3C1, P2RY14, PPP3CB, PTPRC, and RARA) whose suppression strongly rescues eye neurodegeneration. Dysregulated pathways identified from CNS and diMN data characterize different stages of disease development. Altogether, our study provides new insights into ALS pathophysiology and demonstrates how AI speeds up the target discovery process, and opens up new opportunities for therapeutic interventions.


This exciting study gives hope not only to people with a devastating rare neurological disease but to everyone looking to fight back against aging. Indeed, identifying targets that are implicated in both aging and disease is a promising approach that is being actively explored by Insilico Medicine [2]. It demonstrates the possibility of speeding up the therapeutic target discovery process by mining available data and applying sophisticated algorithms to focus on specific genes and pathways. Although the downstream analysis certainly requires in vivo validation, PandaOmics’ predictions regarding various clinically significant properties of its targets minimize the risk of failure.

We would like to ask you a small favor. We are a non-profit foundation, and unlike some other organizations, we have no shareholders and no products to sell you. We are committed to responsible journalism, free from commercial or political influence, that allows you to make informed decisions about your future health.

All our news and educational content is free for everyone to read, but it does mean that we rely on the help of people like you. Every contribution, no matter if it’s big or small, supports independent journalism and sustains our future. You can support us by making a donation or in other ways at no cost to you.

Study: Waist-to-Hip Ratio Predicts Mortality Better Than BMI

A new study suggests that waist-to-hip ratio (WHR) has a more linear correlation with all-cause mortality than either body mass...

Restoring Heart Regeneration With a Metabolic Switch

In a recent article in Nature, researchers have restored cardiac regeneration to adult mice by disabling fatty acid oxidation, discovering...

Human Clinical Trials of NMN for Safety and Effectiveness

In a recent paper, researchers reviewed the literature for human clinical trials that address NMN's safety and anti-aging effects [1]....

Lifespan News – Elon Musk and the Living Forever Curse

On this episode of Lifespan News, Ryan O'Shea ruminates on Elon Musk's statement on living forever being a curse rather...


[1] Pun, F. W. et al. Identification of Therapeutic Targets for Amyotrophic Lateral Sclerosis Using PandaOmics – An AI-Enabled Biological Target Discovery Platform. Front. Aging Neurosci. 14, (2022).

[2] Pun, F. W. et al. Hallmarks of aging-based dual-purpose disease and age-associated targets predicted using PandaOmics AI-powered discovery engine. Aging Β 14, (2022).

About the author
Larisa Sheloukhova

Larisa Sheloukhova

Larisa is a recent graduate from Okinawa Institute of Science and Technology located in one of the blue zones. She is a neurobiologist by training, a health and longevity advocate, and a person with a rare disease. She believes that by studying hereditary diseases it’s possible to understand aging better and vice versa. In addition to writing for LEAF, she continues doing research in glial biology and runs an evidence-based blog about her disease. Larisa enjoys pole fitness, belly dancing, and Okinawan pristine beaches.
No Comments
Write a comment:


Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.