ALS-Targeting Supercomputing Firm Adds CMO

ALS drug discovery in silico? Some computer scientists suspect that identifying potential treatments for ALS may be elementary—by using speed-reading deep learning supercomputers. [Image: IBM Research. CC BY-ND 2.0.]

BenevolentBio named gastroenterologist Patrick Keohane chief medical officer this month. The London-based company disclosed last fall that ALS is in its sights, collaborating with University of Sheffield neuroscientists Laura Ferraiuolo and Richard Mead. BenevolentBio is using deep learning supercomputers to speed read and analyze the scientific literature to identify ALS targets and prioritize potential drugs to target them. The University of Sheffield team is currently evaluating two potential strategies for ALS at the preclinical stageBenevolentBio is the first artificial intelligence company to bring a chief medical officer on board. The move is a key sign that deep learning supercomputers may be beginning to cement its position in the healthcare arena.

Meanwhile, the game’s a foot in the laboratory of Robert Bowser in Arizona to identify new therapeutic targets in ALS. The Barrow Neurological Institute team, with the help of IBM supercomputer Watson, identified 5 new RNA-binding proteins (RBPs) implicated in ALS (See January 2017 news). The proteins included Syncrip (hnRNP Q), a RBP that may help keep muscles and motor neurons connected by regulating the neuronal output of neuromuscular synapses (Halstead et al., 2014; McDermott et al., 2014). The loss of neuromuscular junction integrity is a key cause of muscle weakness and ultimately paralysis in the disease.


Halstead JM, Lin YQ, Durraine L, Hamilton RS, Ball G, Neely GG, Bellen HJ, Davis I. Syncrip/hnRNP Q influences synaptic transmission and regulates BMP signaling at the Drosophila neuromuscular synapse. Biol Open. 2014 Aug 29;3(9):839-49 [PubMed].

McDermott SM, Yang L, Halstead JM, Hamilton RS, Meignin C, Davis I. Drosophila Syncrip modulates the expression of mRNAs encoding key synaptic proteins required for morphology at the neuromuscular junction. RNA. 2014 Oct;20(10):1593-606 [PubMed].

Further Reading:

LeCun Y, Bengio Y, Hinton G. Deep learning. Nature. 2015 May 28;521(7553):436-44 [PubMed].

Aliper A, Plis S, Artemov A, Ulloa A, Mamoshina P, Zhavoronkov A. Deep Learning Applications for Predicting Pharmacological Properties of Drugs and Drug Repurposing Using Transcriptomic Data. Mol Pharm. 2016 Jul 5;13(7):2524-30 [PubMed].


deeplearning disease-als topic-newmethods
Share this: