Sanofi Hands 20 Mature Medicines to Cheplapharm, NVIDIA Backs Verily, Roche Inks $1B Bispecifics Deal with Dualitas

By Bio-IT World Staff

September 30, 2026 | Ginkgo Bioworks builds an autonomous lab at Novo's new Waltham R&D facility; Helix will sequence one million biospecimens for Precure, the Mayo Clinic and Thermo Fisher joint venture; Ginkgo, Apheris, AbbVie and Takeda join antibody developability consortium; Oracle launches Oracle Life Sciences Data Intelligence; and more.

Ginkgo Datapoints and Apheris have launched the Antibody Developability Consortium with founding members AbbVie, argenx, Lundbeck, and Takeda; the consortium remains open to additional pharmaceutical and biotech companies. The Consortium is a new industry collaboration designed to help pharmaceutical and biotech companies predict manufacturability and developability risks earlier by building the field’s largest standardized antibody developability dataset. Existing predictive models have been limited by small, fragmented, and inconsistent datasets, while even large internal datasets are constrained in sequence diversity. The Antibody Developability Consortium was created to address this gap by unifying efforts to produce a standardized, purpose-built, diverse dataset and corresponding AI models at scale. Each founding member will contribute proprietary antibody sequences, with Ginkgo Datapoints filling any remaining capacity from publicly available sources to reach 10,000 antibodies in total. Using Apheris's federated learning infrastructure, members can train, benchmark, and refine AI models using the full consortium dataset, without raw proprietary sequences being exposed to other members. The consortium’s initial dataset is targeted for delivery to members by early 2027. Press release.

Domino Data Lab has announced a strategic partnership with EPAM. The collaboration pairs EPAM’s custom systems engineering expertise with Domino’s platform, helping joint customers deploy GxP-validated AI models and governed enterprise analytics across R&D. Domino’s latest industry research identifies legacy system fragmentation, GxP revalidation demands, and governance hurdles as the primary barriers before AI initiatives can deliver impact. Domino provides the platform layer to bridge these gaps, while EPAM delivers the specialized domain engineering to integrate it into enterprise IT stacks. EPAM’s dedicated integration teams and FDEs work alongside Domino’s FDEs. Together, they embed directly with customer teams to accelerate deployment timelines, modernize legacy code, and ensure strict compliance across heterogeneous IT architectures. Press release.

10x Genomics has begun commercial shipments of Atera. Introduced in April of this year, Atera brings together whole transcriptome analysis, sub-cellular resolution, single cell sensitivity and unprecedented scale, removing longstanding tradeoffs that have constrained the size, scope and accessibility of spatial studies and expanding the range of biological questions researchers can address. Now, Atera is moving into the hands of researchers around the world, from those scaling existing spatial programs to researchers exploring spatial biology for the first time, across discovery, translational and biopharma research. Press release.

Lifebit has launched AI Scientist Studio. The new platform enables researchers to build and deploy custom AI scientist agents within secure Federated TREs, allowing organizations to use frontier AI models on governed health data while ensuring data never leaves the source. Rather than requiring researchers to transfer data into external AI environments, Lifebit brings AI to the data, enabling analysis to take place where governed datasets already reside. Built on Lifebit’s Federated TRE architecture, AI Scientist Studio applies the Five Safes framework to agentic AI. Researchers can bring their own or choose any foundation model, scientific tool and knowledge base, then build specialist agents for cohort discovery, deep phenotyping, analysis and evidence generation. Agents execute on compute inside each data custodian’s governed environment, inheriting user permissions and operating federated across sites. Patient-level data never moves; chat history, outputs and complete audit trails remain within the environment, and only approved results leave through Lifebit’s Airlock controls. Press release.

Cytek Biosciences has expanded its Cytek Cloud digital ecosystem with a new Cloud Analysis tool, establishing the industry’s first end-to-end platform that integrates the entire flow cytometry workflow from panel design to data analysis into one system. Cytek Cloud is built by the scientists and engineers behind Cytek Full Spectrum Profiling (FSP) technology and gives researchers a consistent experience from experiment planning through data analysis while reducing the need to move between systems or recreate work along the way. Additionally, Cytek is introducing the Cytek Cloud Premium license, a new subscription tier that includes all the capabilities of the Pro license, plus cloud-based data analysis and online storage. Press release.

Quanterix’s PhenoCode 2.0 Molecular Barcoding Kit, which allows researchers to barcode antibodies for immuno-oncology and neuroscience applications, is now commercially available for ordering. Two additional offerings are set to ship in the fourth quarter of 2026: a TIME Panel for analyzing the tumor immune microenvironment in tissue samples using PhenoCycler-Fusion technology, and a High-plex ADC Panel that allows simultaneous measurement of antibody-drug conjugate targets and immune cell activity within a single tissue section. Quanterix's Accelerator Laboratory Services are now accepting requests for custom ADC panel development, offering biomarker services from discovery through clinical trial testing under CLIA, GCLP, and ISO 15189 quality standards, intended to give tumor immunology researchers more comprehensive views of tissue biology beyond single-marker analysis. Press release.

The National Institutes of Health's National Library of Medicine launched Linked Discoveries, an experimental tool designed to help researchers understand how individual studies connect to the broader scientific literature. The tool became publicly available on September 24, 2026, and currently covers more than 29 million PubMed publications. It uses artificial intelligence to identify related publications and lets users explore those connections through graph and timeline views, filtering by conditions, genes, and chemicals while surfacing information on citations, reviews, retractions, and NIH-funded research. The tool is intended to give researchers contextual connections among publications that can support replication and reproducibility efforts, though it does not itself judge study quality or confirm whether findings have been replicated. The National Library of Medicine plans to refine the tool based on feedback submitted directly within the platform. Press release.

Deep Origin announced that four organizations—ImmVue, Synko, Sanford Burnham Prebys, and SyzOnc—are now deploying its in silico drug safety prediction models in a pilot program. The models were developed by the PREDICTS consortium (Pharmacological Research and Evaluation through Digital Integration and Clinical Trial Simulation) under an ARPA-H grant of up to $31.7 million awarded to Deep Origin in 2025. The consortium has released 62 of a planned 77 machine learning-based ADMET models, covering absorption, distribution, metabolism, excretion, and toxicity endpoints such as genotoxicity, hERG inhibition, and drug-induced liver injury built on Togo. Press release.

Oracle and the Buck Institute for Research on Aging are collaborating to advance intelligence into age-related disease and support discoveries that can extend the healthy years of life. Buck will leverage Oracle Life Sciences Data Intelligence, which encompasses 122 million+ de-identified longitudinal Oracle Health Real-World Data records and advanced analytics to research biomarkers that aim to enable early detection and targeted intervention for aging-related complications. The research will also be used to pursue qualification for the first FDA-grade "Intrinsic Capacity Score," a vitality score for healthy aging. Press release.

Clean Cells, a GMP biologics quality control CRO, launched a next-generation sequencing (NGS) testing service for viral safety assessment of viral vectors, recombinant preparations such as mRNA and recombinant proteins, and vaccines. Alongside the launch, the company acquired Xegen, a French bioinformatics firm whose PolyPhemeDiag platform provides the analytical software layer for interpreting sequencing data. The new offering is designed to meet ICH Q5A(R2) and European Pharmacopoeia chapter 2.6.41 requirements and addresses limitations of classical adventitious virus assays, which cannot adequately screen viral vectors when no neutralizing antiserum is available. By combining wet-lab testing, sequencing, and bioinformatics under one validated quality system, Clean Cells aims to give sponsors a single, traceable analytical pipeline from sample to interpreted result for viral vector and gene therapy programs. Press release.

Verily Health received new investment from NVIDIA, alongside an additional investment from existing backer CU Healthcare Innovation Fund II. NVIDIA joins existing investors Alphabet, Series X Capital, and UCHealth. The funding will support scaling of Verily's Pre Platform, a data and AI platform for healthcare organizations, and Verily Me, its patient engagement offering. Verily and NVIDIA have collaborated since October 2025 on AI integration, including development of Forecast 1.0, a foundation model that combines genomic and electronic health record data to inform long-term health risk predictions. The investment is intended to support continued development of data infrastructure and AI tools aimed at helping healthcare organizations use combined clinical and genomic information in care decisions. Press release.

The Research Foundation for the State University of New York, working with Stony Brook University, developed asymmetric charged lipid vesicles, or liposomes, for pharmaceutical delivery. Unlike conventional liposomes, the new design allows the inner and outer lipid layers of the vesicle to carry different electrical charges, addressing a limitation in drug payload capacity found in traditional liposomal delivery systems. The technology is intended to increase drug concentration, extend circulation time in the body, and potentially reduce required dosages for chemotherapeutic drugs and other charged biomedicines, with additional applications in DNA delivery involving modification of viral membranes. The invention is covered by patent US20230338288A1 and is available for licensing. Press release.

Regeneron selected InspiroGene, a McKesson business, as a distribution partner for Otarmeni (lunsotogene parvec-cwha), an FDA-approved gene therapy for genetic hearing loss. InspiroGene will manage distribution of Otarmeni to Authorized Treatment Centers across the United States, providing specialized storage, inventory management, and secure distribution as patient access expands. Otarmeni is the first FDA-approved gene therapy for severe-to-profound sensorineural hearing loss linked to biallelic OTOF gene variants; it works by delivering a functional copy of the OTOF gene to the inner ear to restore hearing pathways. Eligible patients must have preserved outer hair cell function and no prior cochlear implant in the treated ear. The therapy holds Fast Track, Orphan Drug, Rare Pediatric Disease, and Regenerative Medicine Advanced Therapy designations from the FDA. Press release.

Sanofi and Cheplapharm announced a strategic partnership under which Cheplapharm will acquire 20 mature medicines and three manufacturing sites from Sanofi, while Sanofi receives a 26.4% equity stake in Cheplapharm. The manufacturing sites involved are located in Csanyikvölgy, Hungary (about 400 employees), Jurong, Singapore (about 100 employees), and Ploërmel, France (about 65 employees). Lovenox/Clexane (enoxaparin), excluding U.S. sales, is among the products included in the transfer. Commercial transfer is set to begin in the first quarter of 2027, with full completion expected by the third quarter of 2027. The deal builds on an existing relationship between the two companies dating to 2014 and is not expected to affect Sanofi's 2026 financial guidance. Press release.

Michigan State University is leading a multi-institutional research team, including Rutgers, Emory, Tulane, University of Colorado Anschutz, University of Michigan, and University of Utah, that received a $4.6 million NIH grant, with potential funding of up to $12.8 million over three years, to develop computational models of women's health. The project addresses the historic gap in clinical trial data on how hormonal fluctuations from menstrual cycles, pregnancy, menopause, and birth control affect drug efficacy and side effects differently in women than in men. Plans include organizing decades of hormone research into a public database, building a standard model of the 28-day menstrual cycle and its organ hormone regulation, incorporating real-world patient data, validating predictions using lab-grown organ tissue models, and developing personalized treatment tools for medications such as metformin and GLP-1 drugs. Press release.

Weill Cornell Medicine and Rutgers Health in New Brunswick received a $3 million, four-year grant from the National Institute of Mental Health to establish the New York-New Jersey Center for Actionable NeuroHIV Biomarkers. Co-directed by Dr. Lishomwa Ndhlovu of Weill Cornell and Dr. Tricia Burdo of Rutgers, the center will provide shared molecular and computational tools for studying how HIV affects the central nervous system, including advanced tissue-culturing techniques such as brain organoids. The work addresses the persistence of HIV in brain reservoirs despite effective antiretroviral therapy, which can still cause cognitive difficulties, behavioral changes, depression, and motor impairment. The center will also provide pilot funding to support junior and early-career HIV researchers at both institutions as they work to identify disease mechanisms and pursue targeted therapies. Press release.

Ginkgo Bioworks will build an autonomous laboratory at Novo's new R&D facility in Waltham, Massachusetts, extending an existing relationship in which Novo previously engaged Ginkgo as an external R&D partner. The lab will use Ginkgo's Reconfigurable Automation Cart architecture paired with its Catalyst software, combining robotic hardware with flexible laboratory capabilities that can integrate with AI agents for protocol writing and other scientific tasks. The installation reflects a broader shift in pharmaceutical R&D toward automated, round-the-clock experimental operations in place of manual lab work. The system is intended to allow scientists at the site to spend more time on analysis and discovery rather than routine bench work as Novo continues to expand its use of AI and robotics in research. Press release.

The Advanced Research Projects Agency for Health, part of the U.S. Department of Health and Human Services, launched SPECTRA (Systems for Phenotypic Evaluation, Clinical Trajectories, Response, and Agency), a research program aimed at transforming autism understanding, diagnosis, and care across the lifespan. The program will integrate biological, clinical, behavioral, and real-world data with computational tools to enable earlier and more accurate diagnoses, identify factors shaping individual health trajectories, and support personalized care approaches. It will focus on analysis of genetic and environmental factors, computational modeling, precision interventions, and accessible technologies for communication and quality-of-life assessment. An estimated one in 31 eight-year-olds in the U.S. has autism spectrum disorder, with associated annual costs projected to reach approximately $461 billion by 2025. Press release.

Dualitas Therapeutics and Roche entered a research collaboration to discover bispecific antibodies for immunology and inflammation diseases. Dualitas will screen more than 300,000 bispecific antibody combinations using its DualScreen discovery platform to identify candidates that exploit proximity biology mechanisms for effects beyond those achievable with conventional therapies. Roche will lead all subsequent development, regulatory, manufacturing, and commercial activities. Dualitas will receive $36.5 million upfront, with potential milestone payments and royalties bringing total deal value up to $1 billion. Founded in 2023, Dualitas has an internal pipeline that includes DTX-102 for rheumatoid arthritis, DTX-103 for allergic disease, and DTX-101 for autoimmune dermatologic and gastrointestinal conditions, with candidates entering clinical trials. Press release.

Insilico Medicine announced Longevity Vaccines, a research program to develop preventive treatments for age-related diseases using programmable RNA medicine. The approach uses circular mRNA in lipid nanoparticles to direct patients' own immune cells to eliminate early-stage disease-causing cells, such as senescent cells and fibroblasts, drawing on the company's Pharma.AI platform and aging foundation models for target selection. The initial focus is immunosenescence, the age-related decline in immune function that reduces vaccine responsiveness and raises infection and cancer risk, with metabolic dysfunction and kidney disease under evaluation for future indications. The program follows Rentosertib, described as the first drug candidate fully designed by AI, which recently showed lung capacity improvements in a Phase IIa trial for pulmonary fibrosis. Press release.

Oncodesign Services and Nexus BioQuest, two preclinical contract research organizations, announced a strategic collaboration to give biotech companies more direct access to specialized preclinical expertise. Nexus BioQuest contributes human-relevant in vitro immunology capabilities using 2D and 3D models, while Oncodesign Services brings three decades of in vivo pharmacology experience, including syngeneic, PDX, CDX, and humanized models. The partnership centers on oncology, inflammation, and autoimmunity, with endometriosis highlighted as an emerging area where combining in vitro inflammatory models with in vivo pathophysiological assessment could support development programs. The collaboration is intended to help biotech teams, particularly those with limited internal resources, more easily identify the right specialist expertise for a given scientific question rather than independently sourcing multiple CROs. Press release.

Palomar Labs spun out Ariadne Bio, its first portfolio company, focused on neuropsychiatric drug development. Ariadne Bio launches with AB-300, a clinical-stage non-hallucinogenic serotonin 2A receptor agonist being developed to treat apathy in Parkinson's disease, a condition with no currently approved therapy that affects an estimated 40% of Parkinson's patients, particularly in mid-to-late stages. The Michael J. Fox Foundation for Parkinson's Research is funding a Phase 1b clinical trial expected to begin in the second half of 2026, with clinical sites in Israel and Austria. In preclinical testing, AB-300 increased motivational behavior by 89% in a dopamine-depletion model compared with control. The company completed a Type B pre-IND meeting with the FDA in 2026. Press release.

Precure, LLC, the joint venture formed by Mayo Clinic and Thermo Fisher Scientific, selected Helix as its laboratory and AI tools provider for a large-scale genomics initiative that will process approximately one million biospecimens. Helix will deliver clinical whole genome sequencing and proteome sequencing for the project, deploying Ultima Genomics' UG200 Ultra Sequencing Platform at scale. The initiative is intended to integrate molecular insights into routine clinical care across Mayo Clinic and other health systems, with the goal of enabling earlier disease detection and more preventive approaches to patient care. The resulting genomic information is intended to both inform patient care directly and create a resource that supports ongoing discovery research, advancing precision medicine toward standard clinical practice. Press release.

Helix also announced it has partnered with Databricks to launch Helix Research Workspaces, a Trusted Research Environment built on the Databricks Data + AI platform. The environment gives approved research partners secure, governed access to computing tools and analysis capabilities for more than 550,000 clinicogenomic records from GenoSphere, which combines genomic data with approximately 13 years of electronic health records, medical data, and pharmacy claims. Press release.

The Edinburgh Genome Foundry and the UK's Advanced Research and Invention Agency announced a three-year partnership, the first Activation Partnership for the University of Edinburgh, to fund an automated laboratory that uses AI for cell design and robotics for cell construction. ARIA-funded researchers will gain access to tailored projects, AI-ready data, and experiments spanning medicine, sustainability, and agriculture, with research focus areas including neurological and viral disease treatments, biomanufacturing, sustainable food systems, and human-machine trust. The partnership is also expected to benefit existing and future industry collaborators working on projects ranging from vaccine development to converting plastic waste into useful products, generating data intended to turn individual experiments into continuous learning systems. The bid was led by researchers from the School of Biological Sciences and Edinburgh Innovations, the university's commercialization service. Press release.

The Research Foundation for the State University of New York has developed several computational and AI platforms that are available for licensing.

They combined an ASD interrogator neural network model with a Synaptic Neuronal Circuit hardware architecture to analyze synaptic dysfunctions linked to Autism Spectrum Disorder and related neurological conditions. The system incorporates retrograde messenger analysis to improve biological accuracy in modeling glutamatergic synapses and can run on GPU-accelerated or ASIC-based hardware for real-time simulations. Press release.
Research Highlighter-MatchMaker connects researchers with funding opportunities and potential collaborators using natural language processing to analyze data through three search methods: researcher name, research topic, or text input such as proposals or abstracts. It integrates publication data from Scopus and funding information from SPIN to deliver personalized recommendations through a searchable portal and automated email alerts. Press release.
GrantsMate is an AI-powered research support platform that consolidates funding discovery, collaborator identification, and institutional policy guidance into a single interface. The system uses machine learning, natural language processing, and retrieval-augmented generation with vector databases, and can be deployed on cloud or on-premises infrastructure with a modular, customizable architecture. It includes a memory engine that personalizes interactions by recalling users' previous queries to reduce inefficiencies that arise when researchers separately search for grants, identify collaborators, and navigate administrative requirements. Press release.

bit.bio, a Cambridge-based cell programming company, partnered with German biotechnology firm myriamed to add human iPSC-derived cardiomyocytes to its drug discovery portfolio. The cardiomyocytes are available in cryopreserved, ready-to-use formats and include ventricular and atrial cell types as well as a dilated cardiomyopathy disease model; the cells exhibit spontaneous beating within seven days of thawing and derive from matched donor backgrounds for comparative studies. Press release.

The University of Melbourne joined the Structural Genomics Consortium by establishing SGC-Melbourne, the consortium's first site in Australasia, led by chemist Professor Megan Maher with partner institutions WEHI, Peter MacCallum Cancer Centre, and Murdoch Children's Research Institute. The site will contribute to Target 2035, a global initiative aiming to develop a pharmacological tool for every human protein by 2035, by generating protein-compound interaction data to train AI models for drug discovery, with a particular focus on rare diseases that collectively affect around one in 12 people. Press release.

The University of Vermont, led by neuroscientist Davi Bock, will coordinate a $19 million, five-year National Science Foundation project in partnership with UC San Diego to create the National Connectomics and Ultrastructure Volume Observatory. UVM will receive approximately $15.68 million and UC San Diego $3.32 million for supercomputing contributions. The observatory will combine electron microscopy, artificial intelligence, and high-performance computing to make advanced brain-mapping and cellular imaging tools, currently available only in specialized labs, accessible to researchers nationwide. Four pilot studies will examine mouse hippocampus changes during learning, retinal circuitry variation across species, a parasite crossing the blood-brain barrier, and the internal structures of single-celled organisms. The project is intended to lay the groundwork for a shared national resource supporting three-dimensional mapping of cellular structures and neural circuits across biology and medicine. Press release.

QIAGEN introduced QIAmini, a compact automated sample preparation system for research laboratories. The system is designed to reduce hands-on processing time by more than 80% compared with manual sample preparation, offering a way to automate work that many labs still perform by hand. QIAmini is the third new sample preparation system QIAGEN has launched in 2026 and extends the company's automation portfolio, which spans small research labs through high-throughput research and clinical testing facilities. The launch is aimed at helping laboratories of varying sizes and throughput needs reduce manual labor and improve efficiency in small-batch sample preparation workflows. Press release.