Thomas Evangelidis

Founder, CEO and CTO
​​AI|ffinity s.r.o.

​​​​​​

Thomas Evangelidis is a distinguished Computational Biophysicist and
Cheminformatician with 16 years of expertise in software development for
Computer-Aided Drug Design (CADD) and protein NMR. He is renowned for his
innovative contributions, particularly the "deepScaffOpt" algorithm, which
significantly enhanced protein-ligand binding affinity prediction and earned

accolades in the D3R Grand Challenges of 2017 and 2018. Additionally, his "4D-
CHAINS" software, designed for the rapid chemical shift assignment of all protein

atoms, was prominently featured in Nature Communications.
In late 2021, Thomas founded "AI|ffinity," where he currently serves as CEO and
CTO. The company’s mission is to create an integrated platform that leverages the

capabilities of NMR, AI, and Cheminformatics to streamline the entire target-to-
lead process, including hit discovery, hit-to-lead, and lead optimization, as well as

protein engineering applications like nanobody design and biocatalysis. AI|ffinity's
rapid 1D NMR screening and 4D NMR protein-drug candidate structure

determination technologies are recognized for their industry-leading speed, cost-
effectiveness, and sensitivity. These attributes make their NMR-AI platform

exceptionally suited for intrinsically disordered and aggregating protein targets,
which constitute more than 30% of the eukaryotic proteome, a unique market
offering.
AI|ffinity is headquartered in Brno, Czech Republic, known as the "NMR hub,"
with additional facilities in Prague, the "CADD hub." As a spin-off from Masaryk

University in Brno, the company benefits from privileged access to the state-of-
the-art biophysical facilities at the Central European Institute of Technology

(CEITEC). This collaboration enables AI|ffinity to engage in joint drug discovery
projects with industrial partners, furthering their research and development
capabilities.

Enhancing Drug Discovery with Nuclear Magnetic Resonance & Artificial Intelligence

In current pharmaceutical R&D settings, the first phase of drug development—
drug discovery—can take up to 5 years, cost around $2.6 billion, and has a low
success rate of less than 12% for new drug discovery. Screening small molecules
to find new drug candidates alone can cost up to $4 million and take 3-6 years.
Ultimately, only one out of approximately 10,000 compounds receives FDA
approval and makes it to market.
At AI|ffinity, we couple nuclear magnetic resonance (NMR) technology with
artificial intelligence (AI), cheminformatics, and computer-aided drug design
approaches into a wholly-owned platform to advance and speed up drug discovery
across various therapeutic modalities. AI|ffinity’s NMR-AI platform stands out
from its competitors through several unique features. It employs hyperpolarized
1D NMR to screen approximately 1500 molecules daily at low concentrations,
which is particularly effective for studying aggregating proteins. The platform
integrates 1D NMR data with generative AI and in silico screening, enhancing the
success rate of predicting new drug candidates. Unlike many other methods, it
can target disordered proteins, which constitute over 30% of the human
proteome, by analyzing small molecules' effects without needing the proteins' 3D
structures. Additionally, AI|ffinity maps interactions of drug candidates on
proteins using advanced 4D and 5D NMR techniques, facilitating enhanced lead
optimization with its Computer-Aided Drug Design software. The platform studies
protein dynamics, offering an efficient and cost-effective alternative to X-ray
crystallography and cryo-EM, and is applicable to proteins that are otherwise
difficult to study. Another advantage is its non-destructive nature, allowing the
same protein sample to be reused multiple times. This versatility extends its
applicability to peptide, enzyme, and antibody engineering, making AI|ffinity’s
NMR-AI platform a comprehensive tool for modern drug discovery and protein
research.