MIT researchers employ machine learning to find powerful peptides that could improve a gene therapy drug for DMD.

Duchenne muscular dystrophy (DMD), a rare genetic disease usually diagnosed in young boys, gradually weakens muscles across the body until the heart or lungs fail. Symptoms often show up by age 5; as the disease progresses, patients lose the ability to walk around age 12. Today, the average life expectancy for DMD patients hovers around 26.

It was big news, then, when Cambridge, Massachusetts-based Sarepta Therapeutics announced in 2016 a breakthrough drug that directly targets the mutated gene responsible for DMD. The therapy uses antisense phosphorodiamidate morpholino oligomers (PMO), a large synthetic molecule that permeates the cell nucleus in order to modify the dystrophin gene, allowing for production of a key protein that is normally missing in DMD patients. “But there’s a problem with PMO by itself. It’s not very good at entering cells,” says Carly Schissel, a PhD candidate in MIT’s Department of Chemistry.

To boost delivery to the nucleus, researchers can affix cell-penetrating peptides (CPPs) to the drug, thereby helping it cross the cell and nuclear membranes to reach its target. Which peptide sequence is best for the job, however, has remained a looming question.

MIT researchers have now developed a systematic approach to solving this problem by combining experimental chemistry with artificial intelligence to discover nontoxic, highly-active peptides that can be attached to PMO to aid delivery. By developing these novel sequences, they hope to rapidly accelerate the development of gene therapies for DMD and other diseases.

Results of their study have now been published in the journal Nature Chemistry in a paper led by Schissel and Somesh Mohapatra, a PhD student in the MIT Department of Materials Science and Engineering, who are the lead authors. Rafael Gomez-Bombarelli, the Jeffrey Cheah Career Development Professor in the Department of Materials Science and Engineering, and Bradley Pentelute, professor of chemistry, are the paper’s senior authors. Other authors include Justin Wolfe, Colin Fadzen, Kamela Bellovoda, Chia-Ling Wu, Jenna Wood, Annika Malmberg, and Andrei Loas.

“Proposing new peptides with a computer is not very hard. Judging if they’re good or not, this is what’s hard,” says Gomez-Bombarelli. “The key innovation is using machine learning to connect the sequence of a peptide, particularly a peptide that includes non-natural amino acids, to experimentally-measured biological activity.”

Read the full article here.