Research

(Classical) Quantum Chemistry

My past research in quantum chemistry has focused mostly on classical quantum chemistry, i.e. quantum chemistry running on classical computers. Here I have worked on new free energy simulation methods for biomolecular systems that include nuclear quantum effects. In this context, I have also developed a new hybrid QM/MM (Quanum Mechanics/Molecular Mechanics) method, called QUASAR, that is compatible with free energy simulations and path integral molecular dynamics simulations.

The new QUASAR method I have implemented in a software package called i-QI, a client for the de-facto standard path integral molecular simulation program i-PI. 

QUASAR QMMM Method
The QUASAR QM/MM method illustrated for a protein-ligand system. In the quantum mechanical (QM) region the ligand, nearby protein atoms, as well as close by solvent molecules are included to improve the accuracy of the computations. The remaining atoms are modeled on the classical level (MM) to reduce the computational demands.

 

Quantum Computing for Chemistry & Drug Discovery

My current and future research interests focus on developing new quantum computing-based methods and tools for quantum chemistry and drug discovery. In particular my focus is methods that can run on near-term quantum computers, also called NISQ (Noisy intermediate-scale quantum) devices.

IBM Quantum Computer
IBM Q System One ©IBM

 

Computer Aided Drug Discovery

In the field of computer aided drug design, one of the topics I have specialized on are ultra-large virtual screenings. Ultra-large virtual screening are virtual screenigns where over 100 million compounds are screened against a given target protein. Here, I have developed VirtualFlow, an open source virtual screening platform that for the first time allowed to carry ultra-large virtual screenings in a routine manner. More information can be found here. 

ULVS
Illustration of the chemical space and ultra-large virtual screenings that access large regions of the chemical space.


Applied Drug Discovery

Regarding the methods and tools that I am developing for drug discovery, I am very interested in applying them to real-world drug discovery projects. To date, I have worked on over one dozen applied drug discovery projects, most of them with collaborators in different countries around the world. Many of these projects were related to cancer, neurodegenerative diseases, or viral diseases.

One of the targets I have worked on is KEAP1, where the target site was the NRF2 protein-protein interface on KEAP1. Details can be found in our paper (Gorgulla, Nature 2020).

Regarding viral diseases, after the pandemic broke out around the world, I have started to work on COVID-19 drug discovery in collaboration with Google, who generously provided us with funding of over 1 million USD in computing time. Further information can be found in our joint paper with Google (Gorgulla, iScience 2021) as well as our COVID-19 project homepage.

SARS-CoV-2 Polymerase Complex
Polymerase complex of the SARS-CoV-2 virus. We have targeted many of the critical protein-protein interfaces of the complex with VirtualFlow, as well as other sites such as the active site or the RNA-binding site. In total I have screened over 40 different target sites of SARS-CoV-2. For each of these sites, over 1 billion compounds were screened in collaboration with Google. These screenings were carried out in approximately one month, demonstrating how efficient the VirtualFlow platform can be deployed in real-world projects.