BLOG

Exploring Amyotrophic Lateral Sclerosis (ALS) with PercayAI’s AI Tools

July 20,2020

Welcome to our newest blog series where we investigate Amyotrophic Lateral Sclerosis (ALS) using our AI knowledge mapping tools. Check back for our next installment, in which we analyze transcriptomics from spinal cord samples collected from patients as well as a mouse model.

PercayAI’s augmented intelligence (AI) tools are designed to allow scientists to explore their areas of research in depth from start to finish, whether this means discovering nuanced insights from a multiomics study, or doing a deep dive on a particular protein that has become part of your new hypothesis. In this series, we’ll explore Amyotrophic Lateral Sclerosis (ALS), a devastating neuromuscular degenerative disease for which there is currently no cure. 

 

Before diving deep into datasets, it is often helpful to take a broader look at the existing research to understand the disease manifestation, susceptibility, molecular mechanisms, and potential therapeutics. This is where PercayAI’s newest single-query exploration tool, BioExplorer™ is especially helpful. Akin to a custom, on-demand, interactive literature review, the tool takes your search term (in our case, “amyotrophic lateral sclerosis”) and scours the literature available in PubMed to identify the most important contextually relevant biological concepts. These concepts are clustered based on their similarity and relationship to one another to form a cohesive knowledge graph. 

BioExplorer™ map of the search term “amyotrophic lateral sclerosis."

 

This map shows: 

  • Disease susceptibility factors in pink

  • Cellular processes involved in ALS in orange

  • Clinical manifestations of ALS in green

  • Other neurological disorders in blue

  • Potential therapeutics in yellow

Disease susceptibility factors

The numerous germline mutations known to cause familial ALS, several biomarkers of the disease, and some potential susceptibility factors were identified within the pink themes (spheres). Among the known genes causing ALS are TARDP (shown in “X-linked dominant inheritance”), SOD1 (shown in “ALS biomarkers”), and PFN1 (shown in “ALS-causing mutations”). Heterogeneous nuclear ribonucleoproteins (hnRNPs, identified within “hnRNP mutations in ALS”) are believed to be drivers of the pathobiology of ALS, as they are found in inclusion bodies of ALS patients and several of the known familial ALS mutations are in hnRNP genes (e.g., FUS, TARDBP). RbBP9 is a biomarker for ALS shown in the “ALS biomarkers” theme.

 

Potential ALS drivers or biomarkers identified in genome wide association studies (GWAS) include DPP6, FGGY, and ITPR2 (shown in “GWAS susceptibility genes”). Check out the other spheres highlighted in pink to explore more susceptibility factors.

Cellular processes involved in ALS

At the cellular level, tissues affected by ALS exhibit a stress response as a result of protein aggregates/inclusion bodies and oxidative stress. The “cell stress/misfolded protein response” and “ubiquitination/sumoylation” themes represent several of these processes. The Rho GTPase cycle is thought to be involved in neuromuscular degeneration in ALS in multiple potential roles, including the regulation of apoptosis that is crucial for motor neuron survival. The RAB GTPases are relevant for their involvement in both neuronal synaptic function and autophagy, an important cellular maintenance function and stress response.

Clinical manifestations of ALS

Symptoms of ALS progression include bulbar weakness/palsy (e.g., facial muscle weakness, difficulty chewing/swallowing, dysphonia) and “locked in syndrome” (loss of voluntary muscle control). At the cellular/tissue level, neurofibrillary tangles and lewy body-like hyaline inclusions are present.

Other neurological disorders

Other neurological/neuromuscular degenerative disorders with some similarities to ALS were identified. These include the neuromuscular disorders muscular dystrophy, spastic paraplegia, and Friedreich ataxia. Alzeimer’s disease, Parkinson’s disease, Huntington’s disease, and multiple sclerosis were also identified and clustered as “neurodegenerative diseases."

Potential therapeutics for ALS

While there is no cure for ALS, the knowledge map highlights a few current and failed therapeutics. Riluzole is a medication currently used for ALS with modest effects; it may delay the requirement for a ventilator or tracheostomy and may extend life by a couple of months. Olesoxime is a putative neuroprotective drug that failed phase 3 clinical trials for extending survival in ALS in 2011. Dexpramipexole is another drug that failed to show efficacy in a phase 3 clinical trial in ALS patients, despite promising preclinical evidence of neuronal stress protection and improvement in mitochondrial function. 

 

Despite past failed therapeutics, a recent publication from researchers in the Miller lab at Washington University St. Louis School of Medicine shows promising evidence of a new drug called tofersen in a phase 1-2 trial. This drug is an antisense oligonucelotide targeting SOD1, one of several genes with known ALS-causing mutations, and was found to effectively lower SOD1 levels in cerebrospinal fluid by 2-33% depending on drug dosage. 


Although this promising new drug would only treat patients with SOD1 mutations, comprising a very small percentage of all ALS patients, the antisense oligonuceotide technology utilized by tofersen could be applied to other ALS targets. With advances in AI-powered research tools, like CompBio and BioExplorer, researchers will be able to more easily identify additional ALS targets to provide hope for an even greater number of patients.

 

Let us know what you think

The findings highlighted in this post just scratch the surface. If you have additional thoughts on these analyses, see something interesting we missed, or have any other feedback on the results from this post, we’d love to hear from you! 

 

Keep an eye out for our next posts, in which we will analyze transcriptomics from spinal cord samples collected from patients as well as a mouse model. Read part two on comparing sporadic and familial ALS using human samples.

Share your impressions

Send us an idea or share your feedback below. We may test it in CompBio, and create and share hypotheses on the blog.*

*We are launching this webpage to share knowledge maps of Coronavirus and to encourage hypotheses generation. We invite you to share your ideas, insights, and feedback (“Feedback”) to help us create hypotheses in connection with Coronavirus therapies, treatments, and solutions. Please click on the links below which cover our Right to Use Feedback, and important Disclaimers. By submitting Feedback, you agree that you have read and agree to these terms. Thank you for your participation in this project.

SHARE YOUR THOUGHTS

  • LinkedIn - Black Circle
  • Twitter

PercayAI

4220 Duncan Ave

Suite 201

St. Louis, Missouri

63110 

+1.314.325.1810

© 2019 PercayAI  

  • LinkedIn - Black Circle
  • Twitter