Astera Institute is looking for a Scientist - Computational Biophysics to join the diffUSE Project. You will develop next-generation protein representations that bridge dynamic structural biology to downstream functional applications, building ensemble-aware models that integrate with protein language models and LLMs.
What You'll Do
- Build ensemble-aware protein representations that integrate PLM and LLM embeddings with experimentally derived structural heterogeneity for functional prediction
- Design, develop, and maintain large-scale bioinformatic pipelines capable of processing and managing complex, high-dimensional datasets
- Fine-tune or architect ML models to capture sequence-structure-function relationships, with a focus on dynamic and conformational features
- Synthesize diverse data sources spanning evolutionary history, binding affinity, allostery, and functional annotations to improve model performance and biological relevance
- Collaborate closely with experimental partners to ground computational representations in real biological measurements and ensure models are continuously refined against experimental ground truth
- Contribute to the broader diffUSE infrastructure, helping establish community-wide standards and tools for dynamic structural biology
What We're Looking For
- PhD in bioinformatics, computational biology, machine learning, or a related field
- Strong understanding of protein structure and function
- Demonstrated experience building large bioinformatic pipelines and managing high-dimensional datasets
- Proficiency in fine-tuning or modifying ML models (e.g., transformer-based architectures)
- Collaborative, team-oriented mindset with the ability to drive research questions from conception to execution
Nice to Have
- Familiarity with protein language models (ESM, AlphaFold, etc.)
Team & Environment
Projects operate like high-velocity startups, with a focus on ambitious goals and matching structure to the problem.
Benefits & Compensation
- Competitive compensation package, commensurate with experience and location. Posted range based on Bay Area location.
