Ahmed Rozza

From 15/03/2025 to 10/06/2025

Paired Teams

Big data for cancer-associated mutations in proteins

Northeastern University, Chemistry and Chemical Biology department

Ahmed Rozza is a research fellow based in Budapest, Hungary, with a passion for understanding how biomolecules function at the atomic level. With academic training in both biotechnology (M.Sc.) and chemical sciences (Ph.D.), he approaches his research through a multidisciplinary lens. His work focuses on applying state-of-the-art computational chemistry techniques – including molecular docking, molecular dynamics, quantum mechanics, and hybrid QM/MM methods – to study the structure and function of complex biomolecular systems. He has a good record of first-author publications in respected peer-reviewed journals and remains dedicated to producing research that deepens scientific understanding and contributes to real-world biomedical applications.

This project aims to investigate the structural and catalytic consequences of the most frequent cancer-associated mutations in the ERK2 MAP kinase protein – a key regulator of cell signaling pathways that control proliferation, differentiation, and survival. Mutations in ERK2 have been implicated in various kinds of cancers. Among them, E322K and D321N are classified as gain-of-function mutations, yet the molecular mechanisms by which these mutations alter protein activity remain poorly understood.

ERK2 mutations often occur in a key docking site – a shallow groove opposite the catalytic site – responsible for binding protein substrates. These substrates typically attach to ERK2 via a short linear motif located within an intrinsically disordered region. Notably, the cancer-related mutations do not completely disrupt these bindings but Instead selectively weaken certain bindings while leaving others unaffected.

To elucidate the impact of the gain-of-function mutations, the project employs state-of-the-art molecular modeling approaches, including molecular docking, molecular dynamics (MD), and quantum mechanics/molecular mechanics (QM/MM), along with Machine learning tools (POOL).  This project seeks to uncover the conformational changes and interaction induced by specific mutations (E322K and D321N).

The major challenge lies in accurately building the workflow for studying EKR2 mutations and capturing the atomic/electronic-level variations that can have profound biological consequences.

The insights to be gained from this project will enhance our understanding of underlying molecular mechanisms responsible for keeping the activity of ERK2 observed in cancer-associated ERK2 variants. In turn, these findings may help inform future strategies for targeted drug design and therapeutic intervention.

Although the project is still in progress, significant steps have already been taken toward the planned objectives. The primary aim is to understand the molecular activation mechanism of ERK2 in the wild type and the most frequent variants.

To date, the preparation of ERK2 crystal structures (PDB IDs: 6OPG and 4GT3) has been completed. Based on the 6OPG X-ray structure, the E322K mutation was introduced to generate the phosphorylated mutant system. Molecular dynamics (MD) simulations were conducted for both the wild-type phosphorylated ERK2 (6OPG-2P) and the E322K-2P systems. Snapshots in which the salt bridge between D147 and K149 was disrupted were selected for molecular docking studies. Docking was performed with a tripeptide substrate containing a serine–proline motif across four systems: wild-type phosphorylated ERK2 (6OPG-2P), phosphorylated E322K (E322K-2P), unphosphorylated E322K (E322K-0P), and wild-type unphosphorylated ERK2 (4GT3). Using Schrödinger’s Glide, 64 conformers of the tripeptide ligand were docked, all yielding the same binding pose. Key interactions between the ligand and ERK2 variants have been identified.

Molecular dynamics simulations (1 μs each) have also been completed using GROMACS for three of the ligand-bound systems: wild-type phosphorylated ERK2, phosphorylated E322K, and unphosphorylated E322K.

Analyses of the structural and dynamical properties of both wild-type and mutant ERK2 are currently underway. Preliminary results suggest that the E322K mutation induces local conformational changes that may influence protein function.

In parallel, high-level QM/MM calculations have been performed on the phosphorylated wild-type (6OPG), unphosphorylated wild-type (4GT3), and unphosphorylated E322K (based on 6OPG) systems to characterize the energies and geometries of the molecular orbitals and electronic structures of ATP and catalytic residues. Optimizing the conditions for the QM/MM adiabatic mapping is underway, so that significant insights into how phosphorylation and mutation impact ERK2 catalytic activity can be obtained.

The anticipated findings will advance our understanding of mutation-driven activation mechanisms and may offer important implications for targeted therapeutic strategies involving ERK2, particularly in cancers where EGFR inhibitors show enhanced effectiveness.

Although the project is still ongoing, the fellowship has already had a measurable impact across several key areas:

• Sound Scientific Validation: The work conducted so far—including molecular docking, MD simulations, and QM/MM analyses—is providing new insights into the structural and functional properties of the most frequent cancer-associated ERK2 mutations. These findings will strongly contribute to a deeper scientific understanding of protein regulation and mutation-driven activation.
• Strengthening Research Collaboration with the US: The fellowship has enabled direct collaboration with a leading research group in Boston, facilitating knowledge exchange, supervision, and feedback that have advanced the project’s scientific depth.
• Building Solid Connections and Partnerships: The research environment has offered opportunities to connect with faculty, postdocs, and researchers in related fields, laying the groundwork for potential future collaborations beyond the scope of this project.

I have established a collaboration with a research group in the Department of Chemistry and Chemical Biology at Harvard University. Through this partnership, we will be working on an exciting project that not only advances our shared scientific goals but also contributes to strengthening transatlantic collaboration between Europe and the United States.

I recently became a member of the Boston Protein Design and Modeling Club (BPDMC)—a vibrant community of computational protein engineers and modelers from both academia and industry. I had the opportunity to attend one of its monthly lectures held at Harvard Medical School, where I engaged with leading experts in the field and gained valuable insights through direct discussions with professionals from top academic institutions and biotech companies.

• Paper Submission for Further Publication: Preliminary results are expected to lead to at least one co-authored publication with the US host institution, reinforcing transatlantic research collaboration.
• Career Advancement: The experience is contributing to my professional growth by expanding expertise in advanced molecular modeling approaches and enhancing my academic profile in preparation for postdoctoral and faculty-level opportunities.
• Expanding Collaboration within the NGI Community: The project’s relevance to computational chemistry and precision medicine aligns well with NGI themes, fostering potential for broader EU-US collaboration in tech-driven healthcare research.