From finding a possible drug to getting it approved by the Food and Drug Administration can take more than a decade and cost more than a billion dollars.

An AI model developed by a group of researchers at CUNY’s Graduate Center has the potential to speed up and lower the overall cost of the pharmaceutical R&D process.

The new model, named CODE-AE, is described in a recently released publication in Nature Machine Intelligence. It can screen novel therapeutic compounds to reliably predict efficacy in people.

In tests, it was also able to find potentially more effective tailored medications for over 9,000 patients. Researchers think that this method will speed up the process of finding new drugs and making them more effective.

To find safe and effective therapies and choose an existing medicine for a particular patient, accurate and reliable predictions of patient-specific responses to a new chemical molecule are essential.

Directly testing a drug’s early efficacy on humans, however, is immoral and impossible. To assess a pharmacological molecule’s therapeutic efficacy, cell or tissue models of the human body are frequently used.

Unfortunately, the way a drug works in a disease model is not always the same as how it works and hurts people. This knowledge gap drives up drug discovery costs and reduces productivity.

This “new machine learning model,” as explained by senior author Lei Xie, “can address the translational challenge from disease models to humans. CODE-AE uses biology-inspired design and takes advantage of several recent advances in machine learning. 

“For example, one of its components uses similar techniques in Deepfake image generation.”

You Wu, a Ph.D. student at the CUNY Graduate Center and a co-author of the research, claimed that the new model can address the issue of not having enough patient data to train a generic machine learning model.

“Although many methods have been developed to utilize cell-line screens for predicting clinical responses, their performances are unreliable due to data incongruity and discrepancies,” Wu added. “CODE-AE can extract intrinsic biological signals masked by noise and confounding factors and effectively alleviated the data-discrepancy problem.”

So, CODE-AE is much more accurate and reliable than state-of-the-art methods when it comes to predicting how a drug will work on a specific patient based only on cell-line compound screens.

The next task for the research team is to establish a method for CODE-AE to accurately forecast the impact of a new drug’s concentration and metabolization in human bodies.

The researchers also pointed out that the AI model might be modified to precisely anticipate adverse medication effects in humans.

Source: 10.1038/s42256-022-00541-0

Image Credit: Getty

You were reading: New Development Can Help Predict Your Body’s Reaction To Novel Drug