Cambridge, UK, September 2017
Intellegens, a provider of specialist Artificial Intelligence (AI) algorithms to the drug discovery and material design sectors, has been awarded an InnovateUK grant to develop an experimental platform to enable integration of its unique algorithms with specialist software and modelling pipelines.
Intellegens’ algorithms are used to add insights to data, where large portions of data can be missing.
Missing data is a key feature of experimental data; for example, through experiments we may know which proteins are affected by a certain drug but with millions of drugs and tens of thousands of proteins the percentage of all known activity is small.
On a small scale a human may look at all this data and see that whenever proteins A and B are active for a particular drug then protein C is also active.
Our algorithms can look at all this data and infer the missing values, leading to a vastly improved dataset to help inform further research and development work.
This improved data quality can significantly reduce time and cost of development processes that can take many years.
The innovation in this project will allow Intellegens to scale this service, currently performed on a manual, case by case basis, by combining (1) our unique AI (2) on-demand cloud-based services, and (3) modern web-based frameworks.
Intellegens is a spin-out from the University of Cambridge that has developed a unique Artificial Intelligence (AI) method for training neural networks from incomplete data. The technique, developed in the Department of Physics, has been applied to drug discovery and material design but as the technique is generic it can be applied to many domains where there is big, incomplete data.