Computational Chemist - Generative Molecule & Protein Design

Roche•Mississauga, ON

4d•$145,300 - $269,800•Onsite

About The Position

At Roche Sequencing Solutions, we are building the next generation of sequencing and diagnostic platforms powered by advanced computation and AI. As part of our Computational Science & Informatics Chapter, you will sit at the intersection of computational chemistry, structural biology, and applied AI to directly impact real-world commercial instruments and R&D workflows. You will join an innovative, collaborative environment where you will: Drive Molecular & Protein Design: Build and run structure-based and ligand-based virtual screening workflows to find new substrates, cofactors, and small molecules for our sequencing and biocatalysis platforms. Enzyme Modeling & De Novo Design: Perform advanced structural modeling and molecular docking simulations on natural and engineered enzyme variants, with a heavy focus on optimizing multi-ligand binding pathways. Lead in silico de novo enzyme design, creating novel protein scaffolds and active sites tailored for multi-ligand binding, optimal catalysis, and enhanced structural stability. Simulate Molecular Dynamics: Explore conformational changes, reaction mechanisms, transition states, and electronic structures using advanced physics-based modeling and quantum chemistry tools. Use Computational Tools: Analyze and optimize biopolymers and biomimetic polymers, leveraging rich internal datasets to design new derivatives and tune properties relevant to sequencing and diagnostic performance. Advance Generative AI Pipelines: Develop and scale generative workflows (including Transformers, GNNs, and diffusion models) for property-guided molecule generation and Computer-Aided Synthesis Planning (CASP). Collaborate for Impact: Partner closely with experimental wet-lab biochemists, protein engineers, and synthetic chemists to turn computational insights into real-world experiments. Pilot Emerging Technologies: Scout and implement cutting-edge tools from the literature and the vibrant local AI ecosystem into production-grade workflows.

Requirements

A Ph.D. Computational Chemistry, Biochemistry, Biophysics, Structural Biology, or a highly quantitative related field or a Master’s degree in Computational Chemistry, Biochemistry with 2 years of related experience or Bachelors with 3 years of related experience.
1+ years of intensive research experience (academic or industry) focused on protein modeling, molecular design, virtual screening, or AI/ML for molecules.
Hands-on experience with docking tools, molecular dynamics packages (such as GROMACS, AMBER, or OpenMM), and quantum chemistry tools.
Proficient in Python (NumPy, SciPy, Pandas, PyTorch) combined with practical experience using RDKit for cheminformatics and developing generative molecular models.
A strong grasp of enzyme kinetics, binding thermodynamics, transition-state theory, and how synthetic modifications alter macromolecular structures.

Nice To Haves

You are a curious, collaborative, and driven scientist who loves combining physics-based modeling with modern machine learning to solve complex biological puzzles.
You explain complex models clearly to diverse teams, value reproducible research, and are excited to co-design experiments that bridge the digital and physical worlds.

Responsibilities

Build and run structure-based and ligand-based virtual screening workflows to find new substrates, cofactors, and small molecules for our sequencing and biocatalysis platforms.
Perform advanced structural modeling and molecular docking simulations on natural and engineered enzyme variants, with a heavy focus on optimizing multi-ligand binding pathways.
Lead in silico de novo enzyme design, creating novel protein scaffolds and active sites tailored for multi-ligand binding, optimal catalysis, and enhanced structural stability.
Explore conformational changes, reaction mechanisms, transition states, and electronic structures using advanced physics-based modeling and quantum chemistry tools.
Analyze and optimize biopolymers and biomimetic polymers, leveraging rich internal datasets to design new derivatives and tune properties relevant to sequencing and diagnostic performance.
Develop and scale generative workflows (including Transformers, GNNs, and diffusion models) for property-guided molecule generation and Computer-Aided Synthesis Planning (CASP).
Partner closely with experimental wet-lab biochemists, protein engineers, and synthetic chemists to turn computational insights into real-world experiments.
Scout and implement cutting-edge tools from the literature and the vibrant local AI ecosystem into production-grade workflows.