Accurate prediction of small molecule solubility using material-sparing approaches is critical for accelerating synthesis and process optimization, yet experimental measurement is costly and many learning approaches either depend on quantumderived descriptors or offer limited interpretability. We introduce Solvaformer, a geometry-aware graph transformer that models solutions as multiple molecules with independent SE(3) symmetries. The architecture combines intramolecular SE(3)-equivariant attention with intermolecular scalar attention, enabling cross-molecular communication without imposing spurious relative geometry. We train Solvaformer in a multi-task setting to predict both solubility (log S) and solvation free energy, using an alternating-batch regimen that trains on quantum-mechanical data (CombiSolv-QM) and on experimental measurements (BigSolDB 2.0). Solvaformer attains the strongest overall performance among the learned models and approaches a DFT-assisted gradient-boosting baseline, while outperforming an EquiformerV2 ablation and sequence-based alternatives. In addition, token-level attention produces chemically coherent attributions: case studies recover known intra- vs. inter-molecular hydrogen-bonding patterns that govern solubility differences in positional isomers. Taken together, Solvaformer provides an accurate, scalable, and interpretable approach to solution-phase property prediction by uniting geometric inductive bias with a mixed dataset training strategy on complementary computational and experimental data.

Artificial intelligence (AI) continued to stay in the news with several high-profile deals this year, as the pharmaceutical industry readily took to adopting AI models to improve drug discovery. But as the field grows in leaps and bounds, many authorities have prioritised the release of new guidelines, frameworks, and regulations to keep pace with these advances. AI applications such as ChatGPT, an Open AI chatbot that understands human speech and produces in-depth writing, have taken the world by storm and expanded the possibilities of using AI. Across all sectors, AI applications are being used to increase efficiency and reduce costs. This is true not only in the case of the general public using applications for creating images or text, but also for pharma companies to improve drug discovery, clinical trial recruitment, and finding new biomarkers.

Check out the on-demand sessions from the Low-Code/No-Code Summit to learn how to successfully innovate and achieve efficiency by upskilling and scaling citizen developers. In an era when biopharma research in drug R&D continues to be costly and slow, and artificial intelligence (AI)-based drug discovery platforms are rapidly growing, Paris-based pharmaceutical leader Sanofi announced its latest massive AI drug discovery deal, this time with startup Insilico Medicine, worth up to $1.2 billion. The research collaboration comes on the heels of several other high-value AI drug discovery partnership announcements from Sanofi, including with Atomwise in August; a partnership expansion with Exscientia last January; and an equity investment in Owkin a year ago. In June, Sanofi's global head of research platforms, Matt Truppo, said that the goal of these AI collaborations is to reduce drug development timelines by "a few years," which in turn brings down costs. According to an Insilico press release, pharmaceutical companies are moving in one of two directions: "Either they are cutting their AI software projects and firing departments, or, like Sanofi, they are doubling down on innovative technology – partnering with leading biotechs to develop new therapeutics using AI."

Too much of it kills you." A new report by Data Bridge Market Research analyses the global Artificial Intelligence (AI) in drug discovery market and forecasts that is expected to reach the value of USD 24,618.25 million by 2029, at a CAGR of 53.3% during the forecast period. The start of the world's first Phase 1 clinical trial of a drug developed from scratch using AI was announced by Insilico Medicine. "At the core of this issue is the complexity of human biology. After decades of molecular biology research, we're lucky if we know 5% of the circuitry of human disease." Just to give you a perspective, this is just the 5% of a simplified view of the brain's circuitry (for more about AI neuroscience news): Owkin's CEO Dr. T Clozel (his parents were the founders of Switzerland's Actelion), is intent on using AI to usher in a new era of drug development, by accessing data at scale with federated learning to preserve patient privacy and data security, and by creating an interpretable AI to answer a broad range of research questions. Last September rapid diagnostic solutions for breast and colorectal cancer from Owkin have been granted approval for use, while this month Sanofi's chief dealmaker (Alban de La Sablière) heads to Owkin as CBO, and all these after Owkin secured $80 million from Bristol Myers Squibb last year and a total raised to over $300 million. The Chicago-based Tempus (@TempusLabs), that specialises in AI and precision medicine and has one of the world's largest libraries of clinical and molecular data, announced this week it raised $275 million through equity from previous investors and debt financing from Ares Management (so far the company has raised over $1.3 billion). And also this month GSK announced that expanded its collaboration with Tempus to improve clinical trial design, speed up enrolment and identify drug targets. "This collaboration will provide GSK with unique insights to discover better medicines and transform drug discovery.

It's a journey that increasingly involves AI and machine learning.

Owkin and Sanofi have announced that Owkin is now a unicorn – a startup valued at more than $1 billion – through a new $180 million investment from Sanofi. Sanofi will take a $180 million equity stake and alongside the investment, Owkin and Sanofi will enter a strategic multi-year collaboration to seek out new cancer therapies using AI. The project will focus on four types of cancer including non-small cell lung cancer, triple-negative breast cancer, mesothelioma, and multiple myeloma. They will use Owkin's predictive biomedical AI models to find new biomarkers and therapeutic targets. Owkin will also build prognostic models to predict how a patient will respond to a particular treatment.

The novel coronavirus, also known as COVID-19 made its first appearance in December 2019 in China. It is believed to have originated from bats or pangolins in the meat markets that occupy the streets of Wuhan. Since then, the number of those infected has increased worldwide as the coronavirus spreads rapidly across the globe. Along with the growing fear rate in the United States with new travel advisories, and more deaths reported in the U.S. Pharmaceutical companies are gearing up to have a new vaccine on the market to combat the virus as the number of COVID-19 cases increase in the United States. The U.S. government is working diligently with pharmaceutical companies round the clock to produce a vaccine to treat the vastly growing coronavirus.

British company Exscientia has been working with several pharmaceutical companies (including Sanofi, GlaxoSmithKline, and Roche), offering its artificial intelligence system to aid the drug discovery process. With the new announcement, Bayer are to back the project with €240 million ($266 million) over the course of three years. The focus of this digital transformation of the medication development process will be on the application of artificial intelligence to speed up the discovery of small molecule drug candidates. The drug candidates will have targets linked to oncology and cardiovascular disease. The deal between the two companies, as PharmaPorum reports, will see Bayer owning the rights to the compounds and Exscientia will receive royalties relating to future sales.

UK artificial intelligence company Exscientia has added another big pharma company to its partner roster, with Bayer seeking to use its platform to find new cardiovascular and cancer drugs. Bayer is pledging up to €240 million ($266 million) in upfront fees, ongoing research funding and clinical milestone payments under the terms of the three-year deal. The collaboration will use AI to accelerate discovery of small molecule drug candidates against targets in oncology and cardiovascular disease, with Bayer claiming rights to the compounds and Dundee-based Exscientia eligible for royalties on sales if they reach the market. Cancer and heart disease are at the forefront of Bayer's R&D focus along with women's health, haematology and ophthalmology. For eight-year-old Exscientia, Bayer joins a growing list of drugmakers who see its AI platform as a way to accelerate drug discovery and improve drug development productivity, potentially trimming years off the current 12 to 15 year cycle from early research to marketed product.

How do you see the current landscape for digital and data science in pharma? The timescale under which we operate today is not the fastest. I bet two years from now the same conversations will probably be going on, just with different faces trying to make the same impact in different companies. You have to ask yourself how much of what we're doing right now is truly impactful vs trying to marginally improve an already inefficient process. What's been even scarier is that we're looking at digital being the utopian cure for everything when I actually think it's the reverse – it's becoming something like an anti-bacterial agent that's developing its own resistance and pitfalls, and I think the companies that are going to win in this space with customers are the ones working on the antidote and scale.