Tech | Visa | Scholarship/School | Info Place

Profluence, powered by Salesforce research and backed by Jeff Dean, uses artificial intelligence to discover drugs

Last year, Salesforce, known for its cloud sales enablement software (and Slack), led a project called ProGen to use generative artificial intelligence to design proteins. The researchers behind ProGen claimed in a January 2023 blog post that ProGen is a research moonshot that, if brought to market, could help discover more cost-effective medical treatments than traditional methods.

ProGen’s research published in the journal Nature Biotechnology shows that artificial intelligence can successfully create 3D structures of artificial proteins. But beyond the thesis, the project didn’t make much sense at Salesforce or anywhere else—at least not in a business sense.

That is, until recently.

Ali Madani, one of the researchers in charge of ProGen, founded a company called Profluu that he hopes will bring similar protein-generating technology out of the lab and into the hands of pharmaceutical companies. In an interview with TechCrunch, Madani described Profluu’s mission as “turning the drug development paradigm on its head” by working backwards to create “customized” treatment solutions, starting with the patient and treatment needs.

“Many drugs—such as enzymes and antibodies—are made of proteins,” Madani said. “So ultimately this is applicable to patients receiving AI-designed proteins as drugs.”

While working in Salesforce’s research department, Madani found herself drawn to the similarities between natural languages, such as English, and the “language” of proteins. Madani discovered that proteins—chains of amino acids bonded together that are used by the body for a variety of purposes, from making hormones to repairing bone and muscle tissue—can be treated like words in a paragraph. Protein data, when fed into the generated artificial intelligence model, can be used to predict entirely new proteins with new functions.

Madani and Alexander Meeske, an assistant professor of microbiology at the University of Washington, are working with Profluent to further advance this concept by applying it to gene editing.

“Many genetic diseases cannot pass [proteins or enzymes] Taken directly from nature,” Madani said. “Additionally, gene-editing systems that mix and match new functions suffer from functional trade-offs that greatly limit their scope.In contrast, Proflute can optimize multiple properties simultaneously to enable custom designs [gene] The editor is perfect for every patient. “

It’s not out of left field. Other companies and research groups have demonstrated feasible ways to use generative AI to predict proteins.

Nvidia released the generative AI model MegaMolBART in 2022, which was trained on a data set of millions of molecules to search for potential drug targets and predict chemical reactions. Meta trained a model called ESM-2 on protein sequences, a method the company claims enabled it to predict the sequences of more than 600 million proteins in just two weeks. DeepMind, Google’s artificial intelligence research lab, has a system called AlphaFold that can predict complete protein structures with speed and accuracy that far exceeds older, less sophisticated algorithmic methods.

Profluence is leveraging massive data sets—one containing more than 40 billion protein sequences—to train artificial intelligence models to create new and fine-tune existing gene editing and protein production systems. Rather than developing treatments on its own, the startup plans to work with external partners to produce the “gene drugs” that have the best chance of gaining approval.

Madani claims this approach could significantly reduce the time and money typically required to develop treatments. According to industry group PhRMA, it takes an average of 10-15 years to develop a new drug from initial discovery to regulatory approval. Meanwhile, recent estimates suggest the cost of developing a new drug ranges from hundreds of millions to $2.8 billion.

“Many influential drugs were actually discovered by accident rather than by intentional design,” Madani said. “[Profluent’s] Capabilities provide humans with an opportunity to move from serendipitous discovery to intentional design of the biological solutions we need most. “

Berkeley-based Profluence has 20 employees and is backed by venture capital giants including Spark Capital (which recently led a $35 million round), Insight Partners, Air Street Capital, AIX Ventures and Convergent Ventures. Google chief scientist Jeff Dean also contributed, lending additional credibility to the platform.

Madani said Profue’s focus in the coming months will be on upgrading its AI models, in part by expanding its training data sets and acquiring customers and partners. It has to act aggressively; competitors including EvolutionaryScale and Basecamp Research are rapidly training their own protein-generating models and raising large amounts of venture capital funding.

“We have developed the initial platform and achieved scientific breakthroughs in gene editing,” Madani said. “Now is the time to scale up and start working with partners to deliver solutions that align with our future ambitions.”

#Profluence #powered #Salesforce #research #backed #Jeff #Dean #artificial #intelligence #discover #drugs

Leave a Reply

Your email address will not be published. Required fields are marked *