You are here: Home >> Executive summary
About us
Who we are and what we do
Executive summary
Executive summary for Equinox Pharma
Executive summary
The platform technology of Equinox Pharma is the program INDDEx, which uses logic-based machine learning to discover novel hits. The methodology is a refinement of the patented support-vector inductive logic programming (SVILP) technique.
The starting information provided by customers is a training dataset (i.e. the biological activity of a series of ligands which have been previously screened against a target). In the first step, INDDEx technology uses machine-learning to generate a set of logic-based rules about the relative positions of structural fragments in the ligands that are responsible for ligand activity. These rules can be easily used by medicinal chemists to understand the mechanism of activity, and help guide the hit-to-lead process.
In the second step, regression is used to combine the rules to yield a filtering model which can then automatically screen databases of millions of available molecules (commercially available and proprietary), and output a list of molecules predicted to have high activity as ligands against that target. As well as activity, INDDEx can be used to predict other properties of a molecule, such as toxicity or ADME profiles.
The key advantages of our methodology are:
- Rules are learnt which can be understood by medicinal chemists, and indicate a route for hit to lead via a synthetic chemistry programme.
- The methodology is well suited to discovering hits which are chemically distinct from the training dataset - a process known as ‘scaffold-hopping’ - providing the opportunity for our clients to file composition of matter patents.
- The methodology can learn from a small or large number of chemically-diverse active and inactive compounds forming the training dataset.
- The process can be iterative and used in the lead optimisation process in conjunction with medicinal chemistry.
Customers of Equinox Pharma employ INDDEx technology in traditional drug-discovery programmes, but they find particular utility in rescuing failed projects, realigning programmes, or in patent-busting exercises.