Our Technology
The Wide Data Problem
Breakthroughs in genetics (The Human Genome Project, Next-Generation Sequencing) have not lead to breakthroughs in patient treatment. The issue is that these technologies do not produce big data, but wide data.
The difference is that with big data you have millions of subjects and a managable amount of information per subject. Wide data is the opposite, you have a few thousand subjects and an entire genome of information for each subject.
Wide data is so hard to analyze that most companies avoid it and rather sequence entire populations. But in order to find even slightly complex genetic interaction in such data, one would need to sequence a large multiple of the entire human population.
Wide Data vs. Big Data
Solution - AI specifically designed for Wide Data
biotx.ai has the only AI specifically designed to make wide data manageable. Using graphs of contextual knowledge about metabolic pathways, protein co-expression and protein-protein interactions, we are able to separate the meaningful features from the noise. This allows us to apply all the benefits of big data to wide genomic data.
This leads to the discovery of previously untedectable complex genetic patterns. These patterns allow 1) for an accurate prediction of disease status and drug response (predictive biomarkers) 2) subgrouping patients for clinical trials (stratification) and 3) establishing drug target linkage (drug development).
Request white paper
Products
Data Preparation
AI Analysis
Certification
Data Preparation
AI Analysis
Certification
Data Preparation
AI Analysis
Certification
Data Preparation
AI Analysis
Certification
Data Preparation
AI Analysis
Certification
Data Preparation
AI Analysis
Certification
Data Preparation
AI Analysis
Certification
Case Studies
Companion diagnostic for accurate prediction of hypercholesterol
We are developing a polygenic biomarker that accurately predicts whether a patient suffers from Familial hypercholesterolemia (FH).
Sanofi was looking for a solution to more accurately predict whether a patient suffers from FH. This Next-Generation Biomarker will be used as companion diagnostic (beyond the pill) by Sanofi in order to give doctors more confidence and security when prescribing highly-efficient anti-bodies to patients with FH. Our Next-Generation Biomarker thus allows FH patients to receive the antibody treatment that will cure their disease.
Next-Generation Biomarkers to manage skin diseases
LEO pharma and the LEO innovation lab are working on a large-scale platform that helps patients manage their skin diseases. We are developing the Next-Generation Biomarkers that fill this platform with content. Our collaboration will allow patients to identify the specific condition they are suffering from and the best treatment to seek out.
Prediction of drug candidate efficacy for Parkinson's disease
In our partnership with Orion, we use our Next-Generation Biomarkers to identify subgroups of patients in neuro-degenerative diseases. Details will be added once the exact cope of the work is finalized.
Core Team
Dr. Jörn Klinger
CEO
Data Science
Oxford, UT Austin
During his graduate studies at Oxford and Texas, Jörn investigated human cognition in order to optimize artificial intelligence. Insights from hands-on research in an indigenous community, neural-networks and agent models led to an outside-the-box machine learning understanding.
Dr. Marco Schmidt
CSO
Biochemistry
FU Berlin, Cambridge
Marco is a biochemist with expertise in chemical synthesis and action of drug molecules. While deeply immersed in several experiments during his Marie Curie Fellowship at the University of Cambridge, he realized that there is great potential of using novel machine learning approaches in drug combinations.
Dr. Charles Ravarani
Head of Data
Genomics
Cambridge (PhD and Postdoc)
As Head of Data at biotx.ai, Charles make use of his academic experience in bioinformatics, including a publication in Nature, as well as his practical knowledge of machine learning and pipeline development, with a strong focus on reproducibility and scalability.