Southwest Research Institute (SwRI) has pioneered a technology designed to screen novel DNA-targeting therapeutics aimed at treating cancer and other diseases. By integrating its advanced 3D drug screening software, Rhodium™, with machine learning techniques, SwRI has successfully predicted the DNA binding affinity and cancer cell toxicity of various drug compounds under development. This breakthrough offers a promising approach to designing more selective and effective cancer therapies, potentially reducing the side effects often associated with traditional chemotherapeutics.
Many chemotherapeutics directly target DNA, which can damage the DNA in healthy cells and cause severe side effects, medical complications, and even secondary cancers. While a number of drug development platforms and machine learning methods virtually screen drugs that target proteins, few methods exist for screening drugs that target DNA. SwRI has now successfully demonstrated a virtual screening application to design more effective DNA-targeting therapeutics to combat different types of cancer and infectious diseases.
“Cancer cells often have damaged DNA repair machinery and replicate much faster than healthy cells, which makes DNA a viable target for selective cancer treatment,” said Tristan Adamson, PhD, a research scientist in SwRI’s pharmaceutical and bioengineering department. “SwRI has successfully developed and validated a powerful technique to use Rhodium for drug development campaigns that directly target DNA in oncology research for leukemia, breast cancer, liver cancer and more.”
More selective drug compounds
Dillon Cao, an SwRI scientist working with Adamson, added, “We want to make drug compounds more selective to target a patient’s DNA for maximum efficacy while at the same time enabling scientists to prune away some of the drug toxicities.”
In an internally funded research project, SwRI scientists developed several machine learning training sets, each serving a role in screening potential DNA-targeting oncology drugs. These models have been validated using statistical tests and correlated with published experimental data.
SwRI scientists accurately predicted the effectiveness of drug compounds against several cancer cell models using the training sets and SwRI’s Rhodium virtual screening tool. They now plan to apply this software in future drug development programs to design next-generation cancer therapeutics.
SwRI used internal funding to develop Rhodium software to provide a computer-aided tool to screen possible treatment methods to combat infectious diseases and chemical warfare agents. SwRI’s machine learning capabilities can evaluate more than two million drug compounds in just a few days to identify compounds with high probability for successful treatment with minimal adverse side effects, claims the company.
By making DNA-targeting drug compounds more selective and reducing toxicities, this technology holds significant promise for improving cancer treatment outcomes.
