Eswar Iyer, MS, PhD, CEO and co-founder at Aikium Inc., describes a tool to overpower the data paucity problem in proteins.
In an interview with BioPharm International, Eswar Iyer, MS, PhD, CEO and co-founder of Aikium Inc., elegantly described the elephant in the room for proteomics: why protein predictions often fail despite the huge strides made in harnessing the power of AI for drug discovery. One of the core reasons predictions often fall short is the failure to account for the conformational flexibility of proteins, compounded by the fact that approximately half of the proteome is partially or completely disordered and lacks stable structures.
“Binding these regions is extremely challenging,” Iyer told BioPharm International.“So what you ideally want for protein prediction models is to have a large number of paired binding interfaces in interaction data sets where you know these proteins bind with their targets and then model based on that. But currently, the data is extremely sparse. What we are doing to solve that is we are taking an experimental approach where the AI designs the binding protein like an antibody, and we take the target, which is disordered, and we can have a trillion different...10 to the power of 13 molecules. Each of them is precisely designed by an AI algorithm, so they are not random.”
Aikium has already signed on four partners—three of which are top cancer hospitals—partly based on successes with GPCR targets, which Aikium used as proof of concept for their technology.
Stay at the forefront of biopharmaceutical innovation—subscribe to BioPharm International for expert insights on drug development, manufacturing, compliance, and more.
Mechanistic Hypothesis of Toxicity: Driving Decision Making in Preclinical Translation
July 16th 2025Mechanistic toxicity hypothesis is essential in guiding decision-making and predicting toxicities during the preclinical stages of drug development. The authors highlight the growing importance of integrating advanced technologies like mass spectrometry imaging into toxicology to enhance preclinical translation, foster innovation in therapeutic development, and ultimately improve drug safety and efficacy.