Researchers at Indraprastha Institute of Information Technology Delhi (IIIT-Delhi), in collaboration with Swedish partners, have launched an Indo Swedish project that uses machine learning, deep learning, and generative AI to design and screen novel peptide therapeutics targeting pneumonia causing pathogens, with the goal of shortening traditional drug development timelines and costs.
Glimpse:
The project, titled “AI-based bio-design for developing peptide therapeutics against pneumonia causing pathogens with experimental validation,” has received funding of INR 52.2 lakh from the Indian side and 2.8 million SEK from Sweden. Led by Prof. N. Arul Murugan as principal investigator, along with co-PIs Prof. G.P.S. Raghava and Dr. Vibhor Kumar from IIIT Delhi’s Department of Computational Biology, the initiative combines computational biology with AI to predict peptide properties such as antimicrobial activity, allergenicity, and toxicity, then generate and prioritize promising candidates for lab validation.
Indraprastha Institute of Information Technology Delhi (IIIT-Delhi) has initiated a significant collaborative research effort with Sweden to tackle pneumonia through advanced artificial intelligence techniques. The project focuses on creating peptide-based therapeutics short chains of amino acids that can be engineered to specifically attack disease causing bacteria with high precision. By integrating machine learning, deep learning, and generative AI, the team aims to overcome the limitations of conventional drug screening, where billions of possible compounds make exhaustive laboratory testing impractical.
Prof. N. Arul Murugan, Professor in the Department of Computational Biology at IIIT-Delhi and the project’s principal investigator, explained that virtual screening powered by supercomputers equipped with large numbers of CPUs and GPUs allows researchers to evaluate vast chemical spaces far more quickly and affordably. AI-based scoring functions help rank peptides accurately according to their predicted effectiveness, safety profile, and low risk of triggering allergic reactions. Once promising candidates are shortlisted, they will undergo rigorous experimental validation, including in-cell assays, in vitro studies, and testing on lung organoid models to confirm their real world efficacy and safety against pneumonia pathogens.
The three year project is expected to identify lead peptide candidates and complete initial experimental validation by its conclusion. If successful, it could significantly reduce the time and expense involved in developing new antimicrobial treatments, while also producing patentable leads that pharmaceutical companies might advance into clinical trials. This approach is particularly valuable in the fight against respiratory infections, where rising antimicrobial resistance continues to pose serious challenges to existing therapies.
The initiative highlights the growing role of computational methods in modern drug discovery, shifting from slow, resource heavy lab work toward faster, data driven bio design. It also strengthens international research ties between India and Sweden in the fields of AI and computational biology.
“The number of possible drug compounds and peptides runs into billions, far more than labs can test practically. Virtual screening with AI makes the process faster and cheaper. We are confident that within three years, we will identify promising peptide candidates against pneumonia pathogens and complete initial lab validation.”
By
HB Team
Related News
