The machine learning market is estimated to be worth a staggering $106.52 billion by the year 2030, according to Market Research Future. Combine this with the rise of consumer interest in artificial intelligence (AI) tools like ChatGPT, the world is beginning to mainstream these technologies.
As this occurs, scientists are developing tools to level-up drug development analytics to create stronger outcomes and potentially save more lives. To add clarity to this, we sit down with Pumas-AI Co-Founder and CEO Dr. Joga Gobburu to answer questions about a product that he and his team developed called DeepPumas TM, which is set to change the future of pharmaceuticals.
What is DeepPumas and how is it changing the pharmaceutical industry?
DeepPumas is a proprietary technology exclusive to Pumas-AI that brings the best of scientific modeling and machine learning together into one tool. Users can unravel relationships in healthcare data taking advantage of known biological and physical laws and machine learning concurrently.
What problems does DeepPumas solve?
DeepPumas is positioned to solve challenges across drug discovery, drug manufacturing, drug development, and healthcare delivery. Key process parameters of drug manufacturing can be identified using DeepPumas to enhance productivity. Similarly, in drug development and healthcare, we are all striving to accelerate access to life-saving precision medicines for patients.
For instance, a pharmaceutical company could detect which patients are likely candidates for their new compound, or a hospital system could individualize diagnosis, prognosis, and treatment trajectories by using DeepPumas to develop precision medicine algorithms using Electronic Health Record (EHR) data on a continuous basis.
Share a few additional examples in how DeepPumas can enable better decision making and individualize predictions.
There are infinitely many examples in the use of DeepPumas for scientific modeling and machine learning, but here are a few more to illustrate its sheer power:
- DeepPumas can boost discovery of new targets by identifying key molecular structure features driving efficacy or toxicity
- DeepPumas can improve drug manufacturing efficiency tremendously by pin-pointing critical process parameters that control product quality
- DeepPumas has been shown to predict patients’ survival probabilities based on early tumor scan data
- DeepPumas can help scientists identify complex interplay between data, such as radiographic scans, genome-wide data, labs, dosing, and clinical outcomes
- With DeepPumas, pharma or regulatory agencies can make more informed decisions around “smart” clinical trials to screen new compounds and therapies
- With DeepPumas, healthcare teams can compare the success chances of a patient on two different treatments; then, they can work to continuously improve the therapy over the course of treatment
How is DeepPumas being used in the development and testing of an artificial blood product for DARPA?
Developing an artificial blood product is very challenging. Numerous factors, like manufacturing, lab testing, and animal experiments, and how they translate into humans make this complex. The scientific world understands some biological and physical laws, but not all. This is where DeepPumas enters the conversation to help fill these gaps and design “smart” experiments.
Additional details on this project can be found here: https://www.medschool.umaryland.edu/news/2023/Artificial-Blood-Product-One-Step-Closer-to-Reality-With-46-Million-in-Federal-Funding.html
Do you think DeepPumas can play a role in helping discover a cure for some of today's "incurable" diseases?
Yes! Let me explain how.
There is no substitution to an in-depth understanding of biology and mechanism of drug action. However, we tend to look at where the light shines. There is an “analysis paralysis” that occurs, resulting in healthcare data overload.
DeepPumas can help scientists to build on their disease and drug knowledge to open potentially unlimited avenues to discover optimal targets for new drug development. We are working with, and currently looking for more, pharmaceutical companies to collaborate with us in this area.
If someone wanted to start using DeepPumas today, what would they need to do?
They would need to send a note to firstname.lastname@example.org to get things started. Given the limitless power of DeepPumas and its disruptive nature, Pumas-AI is seeking opportunities to work with scientists on projects.