Complex diseases like dementia, COPD, cardiovascular or metabolic disease don’t have simple, single gene explanations, so why do we expect our analyses to work one SNP at a time? We need to look deeper at combinations of features to understand why some patient sub-groups have higher disease risks or respond to drugs differently.
We stratify patient sub-groups at unprecedented resolution finding new approaches to address unmet medical need and helping clinicians to select the most effective medicines.
Our revolutionary precisionlife platform applies our proprietary mathematical framework and AI techniques to population-scale genomic, clinical and patient health datasets. We use it to rapidly find all the combinations of genomic, phenotypic and clinical features that define disease risk, prognosis and therapy response in a complex disease population. This enables:
- more precise patient stratification
- identification and validation of novel targets
- systematic analysis of repurposing opportunities
- better, more personalized risk scores
Understanding Disease Mechanism
These analyses provide testable hypotheses of the mechanisms driving disease. In the example below, we can hypothesise that mutations in gene shown below as ⑥ may inhibit the insulin receptor (INSR), which is a key activator of an oncogene PI3K. With this brake on PI3K activation in place it may matter less that the body’s genome surveillance mechanisms are compromised by mutations in BRCA2 as tumors are less likely to form in the first place. This is just one of the 3,045 disease signatures identified using the precisionlife analysis on this basic genotypic dataset.
This analysis took just 4 hours on a 4x GPU machine.
The features are often SNPs – single nucleotide polymorphisms, but can be other structural variants such as CNVs or other data types such as epigenetics, phenotypic, clinical or epidemiological/environmental data). By using combinations of features, we can stratify patients into more accurate sub-groups and evaluate each of the factors driving their different disease risks and/or causes.
Unique combinatorial signatures provide a detailed insight into the causes and highlight potential intervention strategies for complex diseases.
This example shows a 6 SNP disease signature associated with significantly reduced risk of developing breast cancer in a BRCA2 positive population.
The signature occurs in 145 BRCA2 positive people who have not developed breast cancer by the age of 55 and 0 early onset (<40 years) breast cancer patients in a 3,000 patient population. Given lifetime risk & disease onset profiles, about 100 of these 145 women would have been expected to have developed breast cancer.
Detailed Patient Stratification
These communities reveal the specific drivers of disease inside patient sub-groups and allow investigation of the specific pathways that are enriched in each group of patients.
Pharma & Biotech Solutions
We have proven that precisionlife can lead us quickly to novel, validated targets, drug discovery opportunities and better personalized medicines.
We produce detailed in silico, in vitro and/or in vivo biological data packages on all our assets, which include:
- Novel target discovery & validation
- Testable hypotheses for mechanism of action
- Drug-like chemical starting points
- Patient stratification biomarkers
- In vitro phenotypic screening data
Precision medicine needs deeper understanding of all complex chronic diseases, not just oncology.
We’re working with clinicians to build tools to help stratify patients and clinicians for more accurate diagnosis and to personalize the selection of the best drug/s, including:
- Patient stratification biomarkers from complex disease phenotypes
- More accurate combinatorial risk scores for disease
- Identification of drug repurposing opportunities
- Systematic analysis of patient risk factors for prediction of drug response