zax.doi.bio/papers

Research & Publications

Scientific Foundation

doi.bio is built on decades of research in structural biology, machine learning, and bioinformatics. Our platform leverages cutting-edge techniques from the scientific community to accelerate molecular discovery.

Key Research Areas

Protein Structure Prediction

The revolution in protein structure prediction, led by AlphaFold and RoseTTAFold, has transformed our ability to understand protein function. doi.bio integrates these predictions with experimental structures from the PDB to provide comprehensive structural context.

Related Work:

• Jumper, J. et al. "Highly accurate protein structure prediction with AlphaFold." Nature 596, 583589 (2021)
• Baek, M. et al. "Accurate prediction of protein structures and interactions using a three-track neural network." Science 373, 871876 (2021)

AI-Powered Literature Mining

Modern natural language processing and large language models enable automatic extraction of structured knowledge from scientific literature. Our platform uses these techniques to build knowledge graphs connecting papers, proteins, and experiments.

Related Work:

• Devlin, J. et al. "BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding." NAACL (2019)
• Brown, T. et al. "Language Models are Few-Shot Learners." NeurIPS (2020)

Molecular Visualization & Interaction

Interactive 3D visualization of molecular structures is essential for understanding protein function and drug design. We build on established tools while adding AI-powered annotations and collaborative features.

Related Work:

• Rose, A.S. et al. "NGL viewer: web-based molecular graphics for large complexes." Bioinformatics 34, 37553758 (2018)
• Sehnal, D. et al. "Mol* Viewer: modern web app for 3D visualization and analysis of large biomolecular structures." Nucleic Acids Research 49, W431W437 (2021)

Knowledge Graphs for Biology

Representing biological knowledge as interconnected graphs enables powerful querying and reasoning capabilities. Our platform creates dynamic knowledge graphs that connect literature, structures, and experimental data.

Related Work:

• Himmelstein, D.S. et al. "Systematic integration of biomedical knowledge prioritizes drugs for repurposing." eLife 6, e26726 (2017)
• Santos, A. et al. "A knowledge graph to interpret clinical proteomics data." Nature Biotechnology 40, 692702 (2022)

Platform-Related Publications

In Preparation

We're actively developing manuscripts describing:

• Novel approaches to integrating AI-powered literature synthesis with structural data
• Methods for automated protein pocket annotation and druggability assessment
• Collaborative workflows for team-based molecular discovery

Community Contributions

doi.bio builds on the incredible work of the open-source scientific software community. We're committed to giving back through:

• Open-source tools and libraries
• Data sharing and API access
• Collaborations with academic research groups
• Support for reproducible research practices

Databases & Resources

Our platform integrates data from leading scientific databases:

• Protein Data Bank (PDB): 243,000+ experimentally determined structures
• AlphaFold Database: 200M+ predicted protein structures
• PubMed: 35M+ biomedical literature citations
• ChEMBL: Bioactive molecules with drug-like properties
• UniProt: Comprehensive protein sequence and annotation data

Stay Updated

Follow our research progress and platform developments:

• Technical blog (coming soon)
• Research updates on LinkedIn
• Platform changelog and release notes
• Academic collaborations and partnerships

Want to collaborate on research or contribute data? Get in touch.