More than ever, there is a need to develop novel treatments, but doing so is becoming more expensive and time-consuming. The U.S. Food and Medication Administration estimates that the development of a new drug can cost billions of dollars and take up to 14 years. Despite all that work, the FDA reported that just 8% of medications reach the market.
We see a huge opportunity amidst this challenge. This is how we intend to do it.
To target a medication, researchers must discover the biological source of an illness, which is generally a protein. A specific protein, for example, may aid tumour growth or cause inflammation. Next, they hunt for a drug that will attack that target, blocking or increasing its function.
Our deep learning engine, the Ligands-Peptides Repurposing Discovery and Search Engine (LPROSE), sifts through millions of potential small molecules compounds from pharmacopoeia in the Chinese, Indian, African and European traditions to find successful medicines. We can then perform protein docking simulations (using Autodock Vina or Schrodinger) to determine how the molecules will act in the human body. The AlphaFold, AlphaFold2 and DeepFold protein folding simulation software have reduced our analysis time by more than half.
Our software library can also forecast whether the treatment will be effective against the target, how it will influence other sections of the body, its toxicity, and potential adverse effects. This is comparable to the SwissADME system from the Swiss Institute of Bioinformatics.
In the mechanical engineering field where our founders are originally trained, a new invention is simulated in a virtual world before we can proceed with a physical prototype build. Our objective is to provide the pharmaceutical business with the same competitive edge that other industries enjoy. Our company’s process is approximately 100 times faster than high-throughput screening, a standard methodology for automating medicinal compound assessment. It’s a million times faster than a custom synthesis by a medicinal chemist. A few companies emerged recently that follow the same AI drug design methodologies. However, our company solely focused on natural, or plant-derived molecular compounds instead of working in the whole chemical space. This is to lower our analysis complexity, reduce expert manpower requirements and accelerate the drug design process.