Drug discovery is a very long, expensive, and often unsuccessful process. The progress is generally slow, frustrating, and labor-intensive, and usually takes up to five to six years to complete.
In some cases, the average time from discovery to launch for a molecule is 10-12 years, and according to Deloitte, the average cost of the US R&D process in 2018 was $2.168 billion per drug. Around one-third of the cost is spent on the drug discovery phase alone.
Moreover, of the 10,000 molecules initially screened, only ten ever make it to clinical trials. For a compound to enter the Phase I trials, the chance of success is slightly below 10% and has not increased in a decade.
Consequently, the expected return on investment (ROI) from drug development steadily declined from 10.1% in 2010 to 1.9% in 2018. Besides advances in genomics, chemical synthesis, and other molecular biology techniques, only around one-third of the estimated 20,000-30,000 known diseases have adequate treatment.
Therefore, finding ways of improving the efficiency and cost-effectiveness of bringing new drugs to the market is imperative for the health and pharmaceutical industry. Improving discovery and clinical trial success rates are critical for the future of drug development.
Here is where the use of artificial intelligence (AI) and machine learning (ML) techniques comes in. AI solutions can improve the predictability, accuracy, and speed of drug discovery, and ultimately speed up the productivity of the entire R&D process of the drug discovery stage. They can:
- Reduce timelines for drug discovery and make the research process more agile
- Increase the accuracy of predictions regarding the effectiveness and safety of drugs
- Improve the opportunity to diversify drug pipelines
In the past several years, many biopharma companies have been adopting a variety of strategies to integrate AI into the discovery process actively. They established cross-functional teams of biologists, chemists, engineers, informaticians, data scientists, and AI experts to transform the way medicines are discovered, developed, tested, and brought to market.
This post will look at the top 7 drug discovery platforms, driven by artificial intelligence and machine learning.
1. Benevolent Platform
Benevolent Platform is an experimental drug discovery platform, built by the UK-based drug development startup BenevolentAI that uses AI technologies to help find new medicines for several serious diseases, including Parkinson’s disease.
The platform focuses on three key areas, Target Identification, Molecular Design, and Precision Medicine. The Benevolent Platform ingests and analyses biomedical information and combines it with AI deep learning to create a bioscience knowledge graph. It allows their scientists to find new ways to treat disease and personalize medicines to patients.
AtomNet is an AI platform by the US-based Atomwise, that uses AI technology to predict small molecule-protein binding affinities and focusses on identifying potential therapeutics for any disease target. AtomNet is a patented structure made of DL Convolutional Neural Networks for hit discovery and lead compound identification and optimization.
AtomNet predicts potential drug cures using supercomputers, AI, and a specialized algorithm that runs through millions of molecular structures. It learns the three-dimensional features of drug-to-target molecular binding and identifies discriminators. The platform can select hits with critical features such as the ability to cross the blood-brain barrier in a short amount of time with new lead compounds obtained in days, bypassing the need for costly and lengthy high-throughput screening experiments.
Insilico is another platform built by AI drug discovery company Insilico Medicine that uses DL approaches to identify protein targets and design novel lead molecules with specified properties. The platform uses deep generative models, and ML techniques based on neural networks that produce new data objects.
The company recently developed a platform called Generative Tensorial Reinforcement Learning (GENTRL), which for the first time, combines two distinct DL models. One example is AI Imagination, which ‘imagines’ molecules with specific properties using two competing networks: a generator, producing images with selected characteristics, and a discriminator, testing if the output is true or false. Once a target is identified, scientists use these DL algorithms to design molecular structures with desired physical and chemical properties.
4. Ligand Express
Ligand Express is a cloud-based platform by Toronto-based company Cyclica that provides an integrated cloud-based and groundbreaking AI-augmented platform for drug design, off-target profiling, system biology linkages, structural pharmacogenomics insights and drug repurposing based on polypharmacology. In the platform, small molecule drugs are screened against repositories of proteomes to determine polypharmacological profiles.
It identifies protein targets based on the structure, while also determining the drug’s effects on these targets. The approach is based on a DL framework. It can be used to identify unwanted drug interactions, reduce toxic effects, identify repurposing opportunities, and generate new knowledge on disease mechanisms in a shorter period. They recently integrated the ML engine POEM, which provides a better understanding of pharmacokinetics and toxicology to predict potential drug candidates’ behavior.
BERG is an AI-based Interrogative Biology platform that combines patient biology and AI-based analytics to engage the differences between healthy and diseased environments. Built by Boston-based biopharma company BERG, focusing on a biology-based approach to therapeutic discovery, the platform utilizes patient population health data, intending to promote faster discovery and development of treatments, and more effective precision treatments. The patient’s biology drives the platform’s results, guiding the discovery and development of drugs, diagnostics, and healthcare applications.
SpliceCore is a cloud-based drug discovery platform by a New York-based biotechnology company Envisagenics that applies AI to the genetic sequence of patients to discover new therapies. The proprietary discovery platform SpliceCore uses RNA data from patients to accelerate R&D in therapeutics and can identify new ribonucleic acid (RNA) targets and design new drugs to correct RNA splicing errors in cancer and genetic diseases.
NuMedii’s predictive Big Data technology enables the discovery of new uses for medicines with a higher probability of successful clinical development. It discovers and de-risks effective new drugs by translating its predictive Big Data technology into therapies with a higher chance of therapeutic success. It uses hundreds of millions of raw human, biological, pharmacological, and clinical data points. It integrates these data with network-based algorithms to discover drug-disease connections and biomarkers that are predictive of efficacy.