IBM offers free AI tools to boost Covid-19 treatment discovery
by Max Smolaks
Including genome sequencing and drug candidate data
by Max Smolaks 6 March 2020
IBM is offering free access to parts of its software and services arsenal that could help identify new treatments for Covid-19.
include a deep search tool trained on existing Covid-19 knowledge, a
list of 3,000 AI-generated molecules that have potential of being
used in treatment, and a huge genomic sequence repository.
of these are available immediately
to qualified scientists and academics.
The gift of data
IBM is a major player in the healthcare industry, and while the deployment of its Watson cognitive computing engine in healthcare is being met with limited success, one of its applications is drug discovery – and that is the expertise that the company is offering in a time of crisis.
Among the software being given away to researchers is a deep search engine tool that can answer questions posed in natural language, created by ingesting multiple Covid-19 and related research datasets, including those published by the White House. Its knowledge base contained 13,335 documents as of Monday, 3 April – detailing everything from clinical trials to DNA sequences – updated daily by a team based at IBM Research Zurich.
IBM has also opened access to its Functional Genomics Platform, which contains information on millions of genes, including two million recent sequences related directly to the SARS-CoV-2.
And finally, on offer is a list of 3,000 unique molecules that could be used in Covid-19 treatments, discovered using IBM’s AI generative framework. Researchers can filter and compare molecules using a variety of graphs and visualizations, and even import additional molecules of their own.
can take up to 10 years and cost as much as $2.6 billion for a new
drug to reach market,” IBM
stated on the website. “To
deal with new viral outbreaks and epidemics, such as Covid-19, we
need more rapid drug discovery processes. Generative AI models have
shown promise for automating the discovery of molecules. However,
there are many challenges in applying existing generative AI
frameworks to accelerate the design of novel drug candidates.”
the last two years, we have been developing robust generative
frameworks that can overcome these challenges to create novel
peptides, proteins, drug candidates, and materials. We have applied
our methodology to generate drug-like molecule candidates for
Covid-19 targets. Our hope is that by releasing these novel
molecules, the research and drug design communities can accelerate
the process of identifying promising new drug candidates for
coronavirus and potential similar, new outbreaks.”