Impact of AI on the Drug Discovery Process
On average, the drug discovery process is a 10-17 year process, with less than 10% probability of success, and costs $2.6B. By introducing AI to the workflow, enterprises can not only shorten the process to 7 – 12 years with a higher than 60% probability of success but also reduce the cost by 22% – 35%.
Target Discovery
1 – 2 YEARS WITH AI
- Omics data analysis
- In vitro function validation
- In vivo validation (ex. knockouts)
- Literature search
AI USE CASES:
- Drug target identification
- Signaling and metabolic pathway predictions
- Literature mining
Discovery & Screening
0.5 YEARS WITH AI
- Discovery & Screening
- Combinatorial chemistry
- Structure-based drug design
- High throughput in vitro screening
- Ex vivo and in vivo screening
AI USE CASES:
- Test library selection
- Screening optimization (siRNA efficiency prediction)
- Drug repurposing
Lead Optimization
1-2 YEARS WITH AI
- Traditional medicinal chemistry
- Rational drug design
AI USE CASES:
- Chemical design
- Interaction prediction
- Manufacturability
- Pricing prediction
- Drug economics
ADMET (Absorption, Distribution, Metabolism, and Excretion – Toxicity)
0.5-1 YEARS WITH AI
- Bioavailability & systemic exposure
- Formulation
AI USE CASES:
- Chemical property prediction
- Pathway simulation
- Pathology imaging
- Efficacy prediction
Clinical Development
3-5 YEARS WITH AI
- Recruiting
- Trial design
- Clinical responses
- Price estimation
- Manufacturing
- Adverse effect prediction
AI USE CASES:
- Clinical trial design
- Biospecimen analysis
- Medical imaging
- Candidate selection
- Biomarker search
Registration
1-1.5 YEARS WITH AI
- USA (FDA)
- Europe (Country-by-country)
- Japan (MHLW)
AI USE CASES:
- Data analytics
Why AI in Drug Discovery?
AI can identify patterns and connections between genes and targets from complex omics data that are not obvious by humans. NeoPulse’s model explainability and causality mining allows us to build novel biological hypotheses to test.
Identifying CMC issues before they happen in drug production and manufacturing reduces costs and helps optimize the products. Large previous screening data in your laboratory allows AI to learn optimal conditions. Narrowing conditions to test minimizes the time and cost to complete the screening.
AI can use clinical data to forecast patient trajectories and optimize treatments. The prediction can be also used to minimize placebo group in clinical trials, which helps patients and saves costs.
Read up on our latest blogs in the field of Bioinformatics
.bloglist .post-img a:hover:after{opacity: 1 !important;}
Blood test-based active TB prediction model by NeoPulse for affordable diagnostics
November 2, 2021
What are the various applications of Artificial Intelligence in Bioinformatics?
AI is good at finding complex associations between data. It can explore not just individual factors that contribute to the problem but also investigate secondary and tertiary factors that can explain a condition. This ability makes AI particularly useful in Bioinformatics.
October 5, 2021