Insilico Medicine

Insilico Medicine is a biotechnology and artificial intelligence company that develops generative AI technologies for drug discovery, pharmaceutical research, clinical development, and scientific innovation. Founded in 2014 by Alex Zhavoronkov, the company focuses on using machine learning, generative models, biological data analysis, and laboratory automation to accelerate the discovery and development of new medicines. Its technology is designed to help researchers identify disease-related biological targets, generate potential drug molecules, evaluate their properties, and make more informed decisions about clinical development.

The company’s central technology offering is Pharma.AI, an end-to-end platform that connects different stages of pharmaceutical research and development. Rather than focusing on one isolated research task, Pharma.AI brings together specialised applications for biology, chemistry, medicine, and scientific research. The platform is intended for pharmaceutical companies, biotechnology organisations, academic researchers, computational biologists, medicinal chemists, and other professionals working in life sciences. Its purpose is to improve research productivity, support the investigation of complex biological questions, and reduce some of the time and resources required to discover and develop therapeutic candidates.

Pharma.AI is organised around several technology families, including Biology42, Chemistry42, Medicine42, and Science42. Each family addresses a different part of the scientific process. Biology42 concentrates on understanding diseases and identifying potential therapeutic targets. Chemistry42 focuses on designing and optimising molecules with desirable properties. Medicine42 supports clinical development and related medical research, while Science42 provides AI-assisted scientific reasoning and research capabilities. These components can be used separately or as part of connected workflows, depending on the requirements of a research organisation.

One of the main applications within Biology42 is PandaOmics, an AI-powered platform for therapeutic target discovery and biomarker identification. It helps researchers investigate the biological mechanisms associated with diseases and prioritise potential targets for further study. The platform analyses different forms of biomedical evidence, including scientific literature, biological datasets, gene-expression information, molecular data, and disease-related knowledge. By bringing these sources together, PandaOmics can help researchers identify patterns and relationships that may not be immediately obvious when examining individual datasets.

PandaOmics also supports multi-omics analysis. Multi-omics research involves examining different layers of biological information, such as genomics, transcriptomics, proteomics, and other molecular measurements. These datasets can provide complementary information about how diseases develop and how biological systems respond to different conditions. PandaOmics provides tools for dataset exploration, comparison, pathway analysis, and target prioritisation. Its knowledge graph and text-analysis capabilities also help researchers connect findings from scientific publications with biological evidence. This can support the investigation of new therapeutic hypotheses and the identification of possible drug targets.

Another major component of Pharma.AI is Generative Biologics, which focuses on discovering and optimising biological molecules. Biologics include therapeutic substances such as proteins and other complex biological products. Their development can involve several challenges, including designing suitable molecular structures, evaluating their properties, and improving their potential performance. Insilico applies generative AI and computational methods to help researchers explore biological sequences and develop candidate molecules with desired characteristics. These capabilities are intended to support research into new biologic therapies and complement the company’s small-molecule drug discovery technologies.

Within Chemistry42, Insilico provides a collection of generative chemistry and computational tools for designing and optimising potential drug molecules. The platform can generate novel molecular structures based on selected properties and research objectives. It supports different approaches to molecular generation, including structure-based and ligand-based design. Researchers can use these methods to explore chemical possibilities, investigate potential interactions with biological targets, and identify candidate compounds for further evaluation.

Chemistry42 also includes tools for molecular property prediction and optimisation. Drug candidates must satisfy several requirements before they can progress through development. Researchers need to consider factors such as potency, selectivity, solubility, stability, toxicity, and other absorption, distribution, metabolism, excretion, and toxicity characteristics. Insilico’s ADMET and off-target capabilities help researchers examine these properties during the design process. This can support the selection and refinement of compounds before more resource-intensive laboratory experiments are conducted.

The platform also provides Alchemistry, a physics-based computational engine for estimating relative binding free energy. This type of analysis helps researchers compare how strongly related molecules may interact with a biological target. Other Chemistry42 capabilities include molecular dynamics workflows, retrosynthesis prediction, model training, and visualisation tools. Retrosynthesis helps researchers investigate possible chemical routes for producing a proposed molecule, while molecular simulation workflows support the study of molecular behaviour. These capabilities can help computational and medicinal chemistry teams examine potential drug candidates from several perspectives.

Insilico also provides MolSpace, which supports the visualisation and comparison of generated molecular structures. Researchers can use visualisation tools to examine the results of molecular generation and explore relationships between compounds. The platform’s model-training capabilities allow organisations to work with specialised models and research data for particular scientific tasks. These tools are intended to support more customised research workflows, although their practical requirements depend on the available data, computing resources, and scientific expertise.

For clinical development, Insilico offers inClinico, an AI-powered platform designed to predict clinical trial outcomes and support clinical research decisions. Clinical development is a complex and expensive stage of pharmaceutical research. Many potential drug candidates fail during clinical testing because their safety, effectiveness, patient selection, or trial design does not meet the required expectations. InClinico analyses different types of information, including clinical trial protocols, drug characteristics, transcriptomic data, and other relevant evidence, to estimate the likelihood of clinical development success.

InClinico is intended to help pharmaceutical researchers and development teams evaluate clinical programmes and examine factors that may influence trial outcomes. Its predictive capabilities can support decisions about trial planning, programme prioritisation, and the assessment of development risks. However, its predictions should be interpreted as research support rather than guarantees of clinical success. Actual trial outcomes depend on many factors, including patient characteristics, study design, treatment response, safety findings, and regulatory requirements.

Insilico’s Science42 family extends its platform into scientific reasoning and research assistance. One of its key offerings is DORA, a multi-agent generative research assistant designed to help scientists investigate questions and coordinate research tasks. Instead of relying only on a single conversational response, agentic workflows can use specialised tools and models to perform different parts of a research activity. This can help users organise information, explore scientific questions, and work through research processes that involve multiple steps.

The company also provides tools and infrastructure for developing, evaluating, and improving scientific AI models. Its MMAI Gym and related benchmarking resources are intended to support the assessment of model capabilities across drug discovery and development tasks. The platform includes model-related resources, scientific benchmarks, and Model Context Protocol servers that expose selected Pharma.AI capabilities to compatible AI agents. These features are relevant to organisations exploring how specialised biological and chemical tools can be incorporated into wider AI-assisted research workflows.

Another important part of Insilico’s work is Life Star 2, its automated laboratory operating environment. AI-driven drug discovery does not end with generating a biological hypothesis or designing a molecule. Proposed candidates must undergo laboratory experiments and other forms of validation. Life Star is intended to connect computational research with automated experimental processes, helping scientists investigate biological targets, screen compounds, and conduct research activities more efficiently. By connecting software with laboratory automation, Insilico aims to create a more integrated process between digital discovery and experimental science.

Insilico also works on longevity research. Its longevity AI initiatives include PreciousGPT models and the Virtual Aging Cell concept. These projects explore the use of multimodal biological information to understand ageing, estimate biological age, and investigate relationships between ageing processes and disease. Ageing research can involve many interacting biological systems, so the company applies AI to help researchers examine complex datasets and develop hypotheses about biological changes over time. These capabilities support the company’s broader mission of extending healthy and productive longevity.

The company’s research and commercial activities are supported by an internal therapeutic pipeline. Insilico uses its AI technologies to discover and develop potential drug candidates for diseases with significant unmet medical needs. Its programmes cover areas such as fibrosis, cancer, immunology, inflammatory conditions, pain, obesity, metabolic disorders, and other diseases. One of its best-known programmes is the TNIK inhibitor developed for idiopathic pulmonary fibrosis. The programme, associated with the drug candidate rentosertib, has demonstrated how the company’s AI-supported discovery process can contribute to the development of a potential medicine. Its progress through clinical development is an important part of the company’s effort to validate its technology in practical pharmaceutical research.

Other programmes in the company’s pipeline target biological mechanisms relevant to oncology and inflammatory diseases. These include research involving USP1, QPCTL, KAT6, MAT2A, PHD, TEAD, ENPP1, and other targets. Some programmes have been licensed to pharmaceutical partners, while others remain in development or are available for licensing. These arrangements demonstrate that Insilico operates both as a technology provider and as a biotechnology company developing its own therapeutic assets. The stage and commercial status of individual programmes can change as research and clinical development progress.

Insilico has established collaborations with pharmaceutical companies, biotechnology organisations, universities, and research institutions. Its partnerships have involved organisations such as Sanofi, Pfizer, Fosun Pharma, Exelixis, Menarini, Astellas Pharma, and other industry participants. These relationships can involve software access, research collaborations, drug development, licensing agreements, and co-development arrangements. The company’s commercial model therefore includes both the provision of Pharma.AI technologies and the development or licensing of potential medicines discovered using its platform.

The company also supports different business arrangements for organisations interested in its technologies. Customers can obtain access to selected software products or explore broader collaborations involving Insilico’s research capabilities. Its platform is commercially available to pharmaceutical and biotechnology organisations, with subscription agreements and other business arrangements depending on the product and scope of use. This flexibility allows customers to select specialised software access or explore a more extensive research partnership.

Insilico Medicine is intended primarily for professional and institutional users. Pharmaceutical companies can use its tools to investigate therapeutic targets, design molecules, and evaluate clinical development opportunities. Biotechnology organisations can explore AI-assisted discovery workflows and assess potential drug candidates. Academic researchers can use relevant tools to investigate biological mechanisms and analyse scientific data. Computational chemists, medicinal chemists, bioinformaticians, and clinical researchers can also benefit from its specialised capabilities.

It is important to understand that Insilico Medicine is not a general-purpose AI chatbot or a consumer healthcare application. Its products are specialised scientific tools designed for research and pharmaceutical development. Effective use may require knowledge of biology, chemistry, medicine, computational science, or drug development. AI-generated results also require experimental assessment and scientific review before they can be treated as reliable evidence. The platform can support research and decision-making, but it does not replace laboratory validation, clinical trials, regulatory evaluation, or professional scientific judgement.

Insilico Medicine’s combination of generative AI, biological data analysis, computational chemistry, clinical prediction, and laboratory automation gives it a broad role in the life-sciences industry. Its Pharma.AI platform connects several stages of the research process, from identifying a disease-related target to generating potential molecules and examining clinical development possibilities. Through its software products, research partnerships, and internal drug pipeline, the company is working to make AI a more practical part of pharmaceutical innovation.

Pricing

Insilico Medicine does not publish a universal public price list for Pharma.AI. Its software is commercially available through subscription agreements, while larger research collaborations and enterprise arrangements are negotiated according to the customer’s requirements.

• Pharma.AI software access – Subscription: Organisations can obtain access to selected software products or platform capabilities through commercial subscription agreements.
• PandaOmics – Subscription: Access is available through commercial arrangements for research organisations and pharmaceutical customers.
• Chemistry42 – Subscription: Corporate subscription agreements are available, with terms determined by the organisation’s requirements.
• inClinico – Commercial access: Pricing is not publicly displayed and should be confirmed with Insilico.
• Enterprise and research collaborations – Custom pricing: Organisations can combine software access with broader AI research, drug discovery, and development partnerships.

Insilico’s published corporate subscription terms describe typical one-year agreements with fixed fees, although actual prices and contract terms depend on the product and customer. Interested organisations need to contact the company for a quotation.

Review

Insilico Medicine is a specialised AI-driven biotechnology company with a broad set of tools for drug discovery, molecular design, biological research, and clinical development. Its biggest strength is the way Pharma.AI connects different stages of pharmaceutical research instead of concentrating on a single task. The combination of PandaOmics, Chemistry42, Generative Biologics, inClinico, and DORA gives researchers access to capabilities covering target discovery, molecule generation, biological design, clinical prediction, and scientific reasoning.

The platform’s specialised tools are another advantage. PandaOmics supports target and biomarker discovery, while Chemistry42 provides generative molecular design and optimisation capabilities. InClinico adds predictive support for clinical development, and Life Star connects computational discovery with laboratory automation. This range makes Insilico relevant to organisations that want to investigate more of the drug-development process within a connected technology environment.

The company’s internal drug pipeline also provides an important dimension to its technology development. Insilico is not only selling software to pharmaceutical customers; it is also applying its own platform to discover and advance potential medicines. Its clinical programmes and licensing relationships provide practical examples of how its AI research can contribute to drug development. These achievements do not guarantee that every AI-generated candidate will succeed, but they demonstrate the company’s commitment to applying its technology to real therapeutic programmes.

Another strength is its range of research and commercial partnerships. Working with pharmaceutical companies, biotechnology organisations, and academic institutions gives Insilico opportunities to apply its tools to different scientific problems. Its flexible business arrangements also allow customers to obtain selected software products or explore broader research collaborations.

However, Insilico Medicine is not designed for casual users or people looking for a simple AI application. Its tools address complex scientific tasks and may require specialised knowledge in biology, chemistry, computational research, or clinical development. The different modules can also require time and technical experience to understand and integrate into an organisation’s existing workflows.

Pricing transparency is another limitation. The company does not provide a comprehensive public price list for its commercial software, so potential customers need to contact its sales team to understand the cost of access. Organisations may also need to consider data preparation, computing resources, training, and integration requirements when evaluating the total cost of adopting the platform.

AI predictions and generated research outputs must also be independently assessed. A molecule that appears promising in a computational model still needs laboratory testing, preclinical evaluation, and clinical validation. Similarly, a predicted clinical outcome is not a guarantee of what will happen in an actual trial. Researchers should treat the platform as a scientific support system and maintain appropriate experimental and clinical review processes.

Insilico Medicine is best suited to pharmaceutical companies, biotechnology organisations, academic research institutions, and scientific teams that want specialised AI capabilities for drug discovery and development. Its broad platform, research pipeline, and laboratory automation capabilities make it a notable participant in the growing field of AI-driven biotechnology. It is particularly useful for organisations seeking to connect computational discovery with experimental and clinical research.