Posts

AI Creates Novel Proteins Paving the Way for Radical New Therapies!

An exciting partnership between computing and graphics giant Nvidia and AI startup Evozyne has announced that they have been able to produce novel versions of a human protein that has never been seen in nature — but with “enhanced function” and the same safety as native proteins. The researchers say that the AI breakthrough lays the groundwork for potential new therapies for rare disorders.

The collaborators say they developed the novel proteins by “listening to the language of life.”

Natural language processing, or NLP, is a cornerstone of AI programs. NLP is software that is designed to analyze language in any form, from handwritten notes to complex peer-reviewed papers. In healthcare, such NLP algorithms are being written that can analyze any kind of document and other datasets and identify biologically relevant text elements such as the names of genes, proteins, drugs, clinical manifestations of a particular disease, and anything else relevant to a given research team’s target.

Kimberly Powell, vice president of healthcare at Nvidia, says natural language learning techniques can also be applied to genomics, in particular proteins that are encoded by our genes.

“We’ve learned that these large language models can understand relationships just through studying amino acid sequence,” Powell said. “There is information about protein function when you can encode and represent and explore that data in these large language models.”

The results of the collaboration between Nvidia and Evozyne were announced during the recent J.P. Morgan Healthcare Conference in San Francisco. Powell discussed them during a separate briefing with journalists. Neither company develops new drugs, but their technologies are used by biotech and pharmaceutical companies working in drug discovery.

The two companies began working together in 2022, collaborating to develop a new deep-learning model that can learn the rules of protein function. Using those rules, they aimed to design new proteins with improved functions. The model was built on Nvidia’s technology for training and deploying large language models for biology.

When engineering therapeutic proteins, scientists aim to make changes that enhance the protein’s function without compromising its safety. In the research described by Nvidia, Evozyne was able to create a protein with 51 mutations. Despite all of those changes, that protein was still able to achieve a two-and-a-half times enhancement in functionality compared to the native human protein on which it was modeled.

In this case, the protein model was specifically designed to test novel therapeutics for a rare condition known as phenylketonuria (PKU), in which phenylalanine levels build up in the body and cause neurological impairment because individuals with the condition cannot process phenylalanine, an amino acid found in certain foods.

Right now, there are few drugs available to treat PKU, and those with the condition must be on strict diets avoiding foods that contain phenylalanine.

How BigRio Helps Facilitate Investment in AI Startups

Much like the collaboration between Nvidia and Evozyne, BigRio is partnering with startups to develop healthcare solutions with AI at the core of their solutions and our sister company Citadel Discovery was launched in 2021 with the purpose of giving a kind of “open access” to the data and technology that will drive the future of pharma research streamlining and lowering the costs of drug discovery and biological research.

We have launched an AI Studio specifically for US-based Healthcare startups with AI centricity. Our mission is to help AI startups scale and gear up to stay one step ahead of the pack and emerge as winners in their respective domains.

AI Startups face numerous challenges when it comes to demonstrating their value proposition, particularly when it comes to advanced AI solutions for pharma and healthcare. We have taken an award-winning and unique approach to incubating and facilitating startups that allow the R&D team and stakeholders to efficiently collaborate and craft the process to best suit actual ongoing needs, which leads to a faster, more accurate output.

We provide:
• Access to a top-level talent pool, including business executives, developers, data scientists, and data engineers.
• Assistance in the development and testing of the MVP, Prototypes, and POCs.
• Professional services for implementation and support of Pilot projects
• Sales and Marketing support and potential client introductions.
• Access to private capital sources.

BigRio excels in overcoming such initial hurdles, which present nearly insurmountable obstacles to a startup operation.

Rohit Mahajan is a Managing Partner with BigRio. He has a particular expertise in the development and design of innovative solutions for clients in Healthcare, Financial Services, Retail, Automotive, Manufacturing, and other industry segments.

BigRio is a technology consulting firm empowering data to drive innovation and advanced AI. We specialize in cutting-edge Big Data, Machine Learning, and Custom Software strategy, analysis, architecture, and implementation solutions. If you would like to benefit from our expertise in these areas or if you have further questions on the content of this article, please do not hesitate to contact us.

 

/by

Digital Twins Are Reinventing Clinical Drug Trials

A cognitive digital twin is an AI-driven representation of a real-world system. The objects being twinned can be mechanical such as vehicles, ships, and airplanes, or biological, such as organs and biological processes. It is the latter that is radically altering pharmaceutical research and may very well change the nature of clinical drug trials forever.

Traditional drug discovery is a long and complex process that can take years and millions and millions of dollars. The long and intensive process of bringing a new drug through all phases of clinical trials and to market starts with recruiting the right candidates and then proceeds through many steps and phases of testing the drugs vs. placebos in the candidates that have been recruited for the trial.

Finding those patients is one of the most time-consuming aspects of the process. But that is all changing thanks to AI and, specifically, cognitive digital twin (CDT) technologies.

Cognitive digital twins behave virtually the same way, statistically, as their physical counterparts, which makes them ideal for the powerful ability of AI to assimilate massive amounts of data and make remarkably accurate predictions.

Digital twins have been used quite effectively for monitoring health and providing preventive maintenance for some very highly complex systems, such as high-performance sports cars to military aircraft.

Now, they are changing the very landscape of drug discovery by modeling perhaps the most complex system of all organs and even complete human beings. For example, digital twins of patients are now being used to find ideal candidates in that all-important recruitment phase of a drug trial. The twin is created using AI algorithms and machine learning to create a “virtual patient” by leveraging data from previous clinical trials and from individual patient records. The model predicts how the patient’s health would progress during the course of the trial.

This kind of CDT technology is also being used to create “virtual patients” who are “stand-ins” for the control group – the ones getting a placebo – in the typical double-blind drug trial protocol. The digital twin patient predicts how that individual patient would react if they were given a placebo, essentially creating a simulated control group for a particular patient. Think of it as splitting yourself into two distinct exact copies of yourself, one given the actual drug and the other given the placebo as a control. This makes for an even more accurate control group than just splitting all those in the trial into two groups as in typical trials, because the control group is now exactly the same as the group getting the drug. The digital twin virtually eliminates any variance between the drug group and the placebo group that could be based on genetic, physical, and lifestyle differences between the two groups.

Furthermore, replacing or augmenting control groups with digital twins could help patient volunteers as well as researchers. Most people who join a trial do so, hoping to get a new drug that might help them when already-approved drugs have failed. But there’s a 50/50 chance they’ll be put into the control group and won’t get the experimental treatment. Replacing control groups with digital twins could mean more people have access to experimental drugs.

In Silico Research

And finally, another area where CDT technology is making a tremendous difference in drug discovery is in the emerging area of “in silico” research, where digital twinning is used to create so-called “organs on a chip.” Digital twins of the human heart, lungs, and other organs are already being used to hyper-accelerate drug discovery.

One of the promises of CDT is to make complete in silico drug trials from start to finish a reality. Early successes occurring now are paving the way to a time in the not-so-distant future where no humans, nor animals, not even a single living cell will be required for drug discovery — and yet the impact of any given therapeutic or treatment option on a targeted organ, system or even an individual cell can be perfectly charted.

Citadel and AI for Drug Discovery

AI and machine learning are having a tremendous impact on healthcare in America, from streamlining hospital operations, to improved diagnostics and more intuitive telemedicine applications. However, AI’s greatest impact will likely be in the way digital twins and other AI solutions are revolutionizing pharmaceutical research.

To that end, Citadel Discovery was launched in 2021 with the purpose of giving a kind of “open access” to the data and technology that will drive the future of pharma research streamlining and lowering the costs of drug discovery and biological research.

The costs of drug discovery continue to rise, with current estimates exceeding $2 Billion. Not to mention that bringing a drug successfully through all clinical trial phases takes, on average, 10-12 years in research and development. Artificial intelligence and machine learning in drug discovery hold the key to reducing these costs and timelines.

Rohit Mahajan is the President and Co-Founder of Citadel Discovery. He has a particular expertise in the development and design of innovative solutions for clients in Healthcare, Financial Services, Retail, Automotive, Manufacturing, and other industry segments.

Citadel Discovery is dedicated to leveraging AI and MI for the purpose of democratizing access to the data and technology that will drive the future of biological exploration, drug discovery, and health technologies. If you would like to benefit from our expertise in these areas or if you have further questions on the content of this article, please do not hesitate to contact us.

/by

Leading Cancer Center MD Anderson Taps AI for Drug Discovery

One of the country’s leading comprehensive cancer centers has just announced that it is tapping an artificial intelligence-powered drug discovery platform to aid its development of novel cancer therapeutics.

The center is working with AI developer Exscientia to aid its discovery of new cancer drugs. According to an MD Anderson press release, the collaboration will start with “jointly identified oncology targets and then employ Exscientia’s AI platform to design small-molecule drugs.” The resulting candidates will be examined by MD Anderson’s Therapeutics Discovery division and its Institute for Applied Cancer Science, and the most promising prospects will potentially advance into clinical proof-of-concept studies at the Houston cancer center.

MD Anderson’s drug discovery institute, known as IACS, and the cancer center’s other teams have to date helped graduate at least five small-molecule and antibody-based therapies into early-stage clinical testing, including through collaborations with Bristol Myers Squibb, Ionis, Astellas and more.

The financial terms of the joint venture were not disclosed; however, in their announcement, Exscientia and MD Anderson said they will “jointly contribute to and support each program” that is targeted for development.

Exscientia, has been a leader in AI-driven design of large-molecule drugs and antibody therapies. In addition to partnering with facilities such as MD Anderson and well-known pharmaceutical companies, earlier this year, Exscientia found itself with the rights to develop a drug of its own. After wrapping up an AI collaboration with Bayer to develop targets in cancer and cardiovascular disease, the two companies announced that Exscientia would retain the option to develop one of the two targets.

Citadel and AI for Drug Discovery

Similar to the partnership between Exscientia and MD Anderson, Citadel Discovery is sharing knowledge and expertise to better enable drug discovery by providing access to data, models, and results discounted for academics and by developing a sharing platform and an expanded list of drug targets.

Citadel was launched in 2021 with the purpose of giving a kind of “open access” to the data and technology that will drive the future of pharma research streamlining and lowering the costs of drug discovery and biological research.

The costs of drug discovery continue to rise, with current estimates exceeding $2 Billion. Not to mention that bringing a drug successfully through all clinical trial phases takes, on average, 10-12 years in research and development. Artificial intelligence and machine learning in drug discovery hold the key to reducing these costs and timelines.

Rohit Mahajan is a Managing Partner at BigRio and the President and Co-Founder of Citadel Discovery. He has a particular expertise in the development and design of innovative AI and machine learning solutions for clients in Healthcare, Financial Services, Retail, Automotive, Manufacturing, and other industry segments.

CarTwin has leveraged AI and Digital Twin technologies to create a digital, cloud-based clone of a physical vehicle designed to detect, prevent, predict, and optimize through AI and real-time analytics. If you would like to benefit from our expertise in these areas or if you have further questions on the content of this article, please do not hesitate to contact us.

/by