Twist is seeking a Computational Scientist to join their Bioinformatics team to support Antibody Discovery, Development and NGS Analysis. The role will focus on developing cutting-edge computational solutions for protein design, accelerating and scaling Twist Biopharma Solutions' offerings by developing Large Language Model-based solutions for designing novel antibody sequences and therapeutic proteins.
Requirements
- Deep understanding of LLMs and neural network models for biological sequence design, with hands-on experience in model development and deployment.
- Demonstrated experience in antibody and/or other therapeutic proteins for AI-driven discovery and engineering applications.
- Understanding of antibody discovery, affinity maturation, humanization, developability: Deep knowledge of antibody engineering workflows, CDR optimization, species humanization strategies, and therapeutic developability assessment including aggregation, immunogenicity, and stability considerations.
- Proficiency in machine learning frameworks (PyTorch, TensorFlow) and experience with protein language models (ESM, Ablang, ProtTrans, or similar).
- Strong experience in analyzing biological datasets (NGS, experimental assays), data preprocessing for ML, and utilizing common file formats (FASTA, FASTQ).
- Understanding of cloud computing, MLOps workflows, model deployment, and CI/CD pipelines for AI systems.
- Interest in learning and contributing to developing web applications (Django, React) and utilizing database management systems for biological discovery platforms.
Responsibilities
- Lead development of AI/ML-based solutions to design novel antibody sequences using Large Language Models (LLMs) and neural networks for sequence optimization.
- Build and deploy machine learning models for protein engineering, including sequence generation, affinity optimization, and developability prediction.
- Process and analyze NGS data, meta data from wet lab assays (expression, binding data, and developability assessment), and large biological datasets to create training datasets and validate AI-generated sequences.
- Provide technical support and troubleshooting: Debug model performance issues, resolve data pipeline failures, troubleshoot experimental integration problems, and provide day-to-day operational support to ensure smooth team workflows and project continuity.
- Develop software solutions for storing, querying, processing, and visualizing biological data and model outputs.
- Utilize advanced data science techniques and frameworks, including deep learning (PyTorch, TensorFlow), analysis (pandas, numpy, scikit-learn), and visualization (matplotlib, plotly).
- Follow and establish software development best practices for Bioinformatics systems, including model versioning, experiment tracking, deployment pipelines, and code management.
Other
- PhD in Computational Biology, Machine Learning, or related scientific discipline with a minimum of 2-5 years related experience (industry experience preferred).
- Strong communication skills and a balanced ability to work independently and as a team member are desired.
- Collaborate with internal members throughout all phases of Biopharma R&D workflows, translating biological requirements into machine learning solutions and providing technical support.
