Independent education resourceInformation here does not replace care from a qualified health professional.
Peptide Therapy GuideClear peptide education

Educational guide

Insilico Medicine launches Pharma AI Spring Kickoff 2026 webinar

As the AI era becomes increasingly shaped by foundation models, the pharmaceutical industry is entering a new phase of opportunity for discovery, design, and decision-making driven by AI for science. To explore these advancements, Insilico Medicine (03696.HK),

Written by Peptide Therapy Guide Editorial Team
For education only

This guide cannot diagnose a condition or recommend a personal treatment plan. Discuss medical questions with a qualified professional.

As the AI era becomes increasingly shaped by foundation models, the pharmaceutical industry is entering a new phase of opportunity for discovery, design, and decision-making driven by AI for science. To explore these advancements, Insilico Medicine (03696.HK), a clinical-stage generative AI–driven drug discovery company, today announced that the Pharma.AI Spring Kickoff 2026 will be held at 10:00 AM ET on April 14, with registration and event details available at: https://insilico.zoom.us/webinar/register/WN_h7tujok6SdmfDWzkZwRgNg.

The 2026 season of the Pharma.AI webinar series will showcase the ongoing AI revolution in life sciences, including the increased interest in the use of foundation models why specialized models remain essential for biology, chemistry, and translational research; How Pharma.AI brings together foundation models and scientific AI agents within a unified AI-driven workflow for drug R&D and scientific research; and how Insilico's leading "AI trains AI" approach may enable foundation models to be better adapted for scientific and drug discovery applications, accelerating the evolution of AI decision-making systems.

More specifically, the upcoming event will highlight new capabilities across the Pharma.AI ecosystem, including the MMAI Gym for Science, updates to core modules such as PandaOmics, Generative Biologics, and Chemistry42.

As we kick off 2026, our focus is on moving beyond simple AI-driven toward a truly AI-decision ecosystem. With the introduction of the continued evolution of Pharma.AI, we are building the foundation for pharmaceutical superintelligence systems that can reason more effectively, adapt to real scientific workflows, and generate meaningful impact across drug discovery and development. The upcoming webinar brings together exciting new updates and is designed to provide researchers with the latest tools and best practices for tackling the most challenging problems in human health." Alex Aliper, PhD, President, Insilico Medicine

Highlights at a glance

  • MMAI Gym: Turning Foundation Models into High-Performance Drug Discovery Engines

The MMAI Gym for Science, a foundation model training framework, was introduced by Insilico in January 2026. Leveraging over 1,000 drug R&D benchmarks and approximately 120 billion tokens of public and proprietary drug discovery data, the framework utilizes multi-task fine-tuning and reinforcement learning to significantly enhance the performance of foundation models across specialized tasks in drug discovery.

Validating the power of this framework, we demonstrate that MMAI-trained foundation models achieved up to 10X performance gains on key drug discovery benchmarks compared to general-purpose foundation models, which fell short on approximately 75–95% of tasks. Moreover, in March 2026, Insilico and Liquid AI jointly delivered LFM2-2.6B-MMAI (v0.2.1), the first model trained through their first MMAI Gym collaboration. Despite its lightweight, on-premise design, the model delivered SOTA performance across several key tasks. The paper detailing the training process and final performance was accepted at ICLR 2026.

During the upcoming event, attendees will learn how this supervised fine-tuning (SFT) and reinforcement fine-tuning (RFT) training and benchmarking system can significantly improve the performance of causal LLMs on real-world drug discovery tasks, and how to access the platform.

  • PandaOmics: Target Prioritization with Single-Cell and PandaClaw

PandaOmics is Insilico Medicine's AI-driven platform for therapeutic target discovery and indication expansion. It integrates and analyzes large-scale multi-omics and biomedical datasets to help researchers to identify and prioritize disease-specific drug targets and to expand the therapeutic indications of targets of interest.

Recent upgrades to PandaOmics include the incorporation of comprehensive single-cell datasets, which provide enhanced resolution for target identification. In addition, PandaClaw, an agentic AI tool that allows scientists to conduct complex, real-time multi-omics analyses, generate research hypotheses, and perform target evaluations via a simple natural-language interface.

  • Chemistry42: Multi-Target and Advanced Alchemistry

Chemistry42 is Insilico Medicine's AI-driven platform for designing and discovering novel small molecules. It combines generative model ensembles and advanced physics-based methods to help researchers create and optimize novel compounds. A core part of Chemistry42 is Nach01, an AI model trained on billions of data points to understand both natural and chemical language, enabling hundreds of professional tasks and laying the groundwork for a "prompt-to-drug" future.

The latest updates include multitarget support for molecule generation, enhanced results visualization for smoother analysis, Nach01-MMAI for molecule generation, and new Absolute Binding Free Energy (ABFE) calculations in Alchemistry.

  • Generative Biologics: Cyclic Peptide Design & Linear Peptide Optimization

Generative Biologics is a cutting-edge biologics engineering platform. It uses advanced multi-parameter optimization to tackle complex challenges in the design of antibodies, peptides, and other biologic drugs. Powered by more than 10 generative and predictive models and enhanced by precise physics-based tools, Generative Biologics enables the rapid creation of diverse, optimized biologics, allowing scientists to generate viable binder candidates in less than 72 hours.

The platform now includes major updates for peptide design. It introduces a completely new workflow for cyclic peptides, supporting both head-to-tail and disulfide-bond architectures, generating hundreds of candidates in just hours with AI- and physics-based prioritization. In parallel, researchers have successfully optimized linear peptides using the platform to refine the lead candidate, P3, against GLP-1R and to produce dozens of new candidates, with the top variant, P3-1, achieving a sixfold improvement over the original lead.

Pharma.AI is an end-to-end AI platform for drug discovery and development, integrating target discovery, generative chemistry, biologics design, and predictive clinical modeling into a unified AI-driven workflow for pharmaceutical R&D. We hope to see you at our first event as we kick off 2026.

Date: April 14, 2026

Time: 10:00 AM ET

Connected reading

Helpful context for this guide

Source-derived material selected through this article’s indexed topics.

Related questions

01What is nisin?

Some bacterial species produce antimicrobial peptides known as bacteriocins that have been used in the food industry as preservatives. For example, nisin, which is produced by Lactococcus lactis, has broad-spectrum bactericidal activity and has been used as a food preservative throughout the world. Nisin is effective in controlling Gram-positive bacteria such as Clostridioides difficile. In combination with other compounds like ethylene diamine tetra-acetic acid and cinnamaldehyde, nisin has been effective in controlling enterotoxigenic Gram-negative bacteria such as Escherichia coli. Previous studies have used chicken and mouse models to demonstrate the in vivo efficacy of nisin on the microbiome, whereas nisin efficacy has been proven in ex vivo experiments on the human microbiome. To date, no studies have assessed the in vivo effects of nisin in large mammals.

Source: www.news-medical.net ↗
02What are functional peptides?

Conventional pharmacological studies on spices have traditionally focused on secondary metabolites like polyphenols, alkaloids, and terpenes. More recently, food science research has also examined spice proteins and their enzymatic hydrolysates, using proteomic methods such as liquid chromatography–tandem mass spectrometry (LC-MS/MS) to identify short bioactive peptide sequences released from larger precursor proteins.6 Once released during food processing, fermentation, or gastrointestinal digestion, these functional peptides can act as metabolic regulators, antimicrobials, or antioxidants.1 Functional peptides refer to specific protein fragments that, once released from their parent proteins, exert biological activities.1,2 In the context of foods, these activities are most often demonstrated using in vitro biochemical or cell-based assays, and their physiological relevance depends on bioavailability and dose.2 Unlike intact proteins, which can have the potential to be allergenic or difficult to absorb due to their complex tertiary structures, functional peptides may exhibit improved bioaccessibility, and some small peptides can cross the intestinal epithelial barrier via peptide transport systems. However, absorption efficiency varies substantially by peptide sequence and digestive conditions.6 Nutriomics and mechanistic investigations have established that the bioactivity of a peptide is dictated by its physicochemical properties, particularly its amino acid composition, molecular weight, and net charge. For example, the presence of hydrophobic amino acids like proline, leucine, and valine often correlates with high antioxidant and enzyme-inhibitory activity.2,3 Smaller peptides, typically those less than three kilodaltons (kDa) in size, exhibit greater stability against proteolytic degradation in the gastrointestinal tract.3 Moreover, cationic peptides are particularly effective as antimicrobial agents through their electrostatic interactions with bacterial membranes.3

Source: www.news-medical.net ↗
03What was this study about?

It has been noted in around 20 percent of the world population suffers from some form of pain or the other. In many individuals, pain may be relieved initially with pain medications, but soon tolerance develops, and there is a decrease in the efficacy of pain relievers. One of the main symptoms of IBS seen commonly in many sufferers is chronic abdominal pain. Professor Lewis said, "All pains are complex, but gut pain is particularly challenging to treat and affects around 20 percent of the world's population. Current drugs are failing to produce effective pain relief in many patients before side effects limit the dose that can be administered." Professor Brierley echoed this statement saying, "Internal organs have a complex network of sensory nerves that have a wide array of voltage-gated ion channels and receptors to detect stimuli... The hypersensitivity of these nerves in disease often contributes to the development of pain."

Source: www.news-medical.net ↗
04How do these peptides act?

These peptides, like the parent compound AC253, acted as antagonists at the AMY receptor. They were also resistant to protein breakdown, and crossed the blood-brain barrier easily when injected into the abdominal cavity, to localize in the hippocampus, which is crucial in memory. These peptides protected the brain against beta-amyloid injury, and normalized the AD-associated impairment of the memory-associated long-term potentiation of nerve impulses in the hippocampus. They improved memory testing results, and reduced the level of inflammation in the brain. These effects appear to be mediated via the blockade of AMY receptors. For instance, inhibition of microglial AMY receptors reduce the activation of the inflammasome NLRP3. This reduces the secretion of inflammatory chemicals in the surrounding brain tissue, which offers another mechanism for lower amyloid production. In addition, these peptides increase the rate of outflow of amyloid beta from the brain, which also contributes to a lower level of amyloid after treatment. These marked changes all occurred within a relatively short span of treatment. A very important additional finding was that treatment with these peptides brought about improvement in mice which were showing signs of well-established AD in the brain as well as in their behavior. This is unique in that most therapies fail to affect the progress of AD once it has begun to manifest clinically. Peptides also have fewer off-target effects. Small molecules are easy to administer, inexpensive to make and cross the blood-brain barrier more rapidly. For this reason, the team resorted to computational tools and artificial intelligence to come up with a new small molecular drug based on these peptides. This can be taken orally, and is similar in size and structure to the medications used for medical conditions like high blood pressure. An optimized version is being developed to enable human trials to be conducted. The work so far has taken about two decades, building step upon painstaking step to come up with the right solution. However, says Jhamandas, “Occasionally you come across a discovery that has the potential to change the game in a very fundamental way, like hitting a home run, and I'm very excited that we are really on to something here.” Short amylin receptor antagonist peptides improve memory deficits in Alzheimer’s disease mouse model. Rania Soudy, Ryoichi Kimura, Aarti Patel, Wen Fu, Kamaljit Kaur, David Westaway, Jing Yang & Jack Jhamandas. Scientific Reports, volume 9, Article number: 10942 (2019). https://doi.org/10.1038/s41598-019-47255-9. https://www.nature.com/articles/s41598-019-47255-9

Source: www.news-medical.net ↗
05What is the concept of the immune self, and how has it evolved over the decades?

Adaptive immunity is the ability of specific lymphocytes to differentiate between self and non-self (foreign) antigens and defend the body by selectively destroying non-self-peptides. This concept is possibly the most crucial factor in several immunological medical domains and is increasingly being explored across cancer immunotherapy, vaccine design, pathogen identification, and autoimmune disorders (including allergies). A growing body of literature elucidates the importance of peptides, short amino acid chains linked via peptide bonds, in providing the adaptive immune system with the information required to effectively distinguish between self and non-self particles. This has resulted in the proposal of the ‘immune self’ concept, which postulates that self-similarity is a fundamental determinant of immune recognition. First introduced by Frank MacFarlane Burnet in 1949, the immune self-concept and its sister, the self-nonself theory, have substantially evolved over the decades. Initially driven by observations from Medawar’s early transplantation experiments, Nils K. Jerne (1974; eigen-behavior theory), Polly Matzinger (1994; danger theory), and most recently, evidence from research conducted independently by Waldmann, Mitchison, and Janeway has refined the immune self-concept from ‘all body elements are self, and foreign elements are non-self’ to the most recent ‘infectious non-self (foreign and usually harmful) versus noninfectious self (safe) elements.’

Source: www.news-medical.net ↗
comparison

Comparisons

Side-by-side pages for commonly compared peptides and research compounds.

Source: peptideuniv.com
Research context

Read sources and limitations before applying a claim.

Longevity, Performance & Obesity Research

A research peptide formulation developed to investigate metabolic regulation, mitochondrial function, and nutrient-sensing pathways.

Source: mypeptidematch.com ↗
P

About the author

Peptide Therapy Guide Editorial Team

Editorial team for Peptide Therapy Guide.

View all articles →