Building and Deploying AI in Life Sciences R&D: Proteins, Genomics and Imaging
For bench and computational scientists, core-facility staff and research IT teams who want to run and critically judge AI models for protein structure and design, genomics and biological imaging.
The curriculum is being prepared and reviewed. Enrollment is not open yet.
About this course
AI models are now part of routine life-science research, but each family comes with its own failure modes and fine print. This course starts with structure prediction: how AlphaFold 3, Boltz-2, Chai-1 and OpenFold3 predict complexes, how to read pLDDT, PAE and ipTM, and when a confident prediction is still wrong. It moves on to generative protein design with RFdiffusion3, ProteinMPNN and BoltzGen, and to protein language models such as ESM, before turning to genomics: variant calling with DeepVariant and NVIDIA Parabricks, variant-effect prediction with AlphaMissense and AlphaGenome, DNA foundation models such as Evo 2, and single-cell foundation models.
The imaging module covers cell and nucleus segmentation with Cellpose-SAM and micro-SAM in napari, medical and preclinical imaging with MONAI, and AI-assisted cryo-EM model building. The final module is about doing this responsibly and at scale: licenses that restrict commercial use of weights or training data, running models on GPU clusters with containers and Slurm, reproducibility and reporting, and dual-use and biosecurity obligations including nucleic-acid synthesis screening. Model versions, licenses and availability are stated as of October 2026.
What you will be able to do
Run a biomolecular complex prediction and interpret pLDDT, PAE and ipTM to decide how far to trust it
Choose between AlphaFold 3, Boltz-2, Chai-1 and OpenFold3 given license, input types and compute constraints
Describe a generative protein-design pipeline from backbone generation to sequence design and in-silico filtering
Select appropriate AI tools for variant calling, variant-effect prediction and regulatory-genomics questions, and state their limits
Segment microscopy images with a generalist model, fine-tune it on lab data and validate the result quantitatively
Plan GPU, container and storage requirements for running these models on a shared research cluster
Apply licensing, reproducibility and biosecurity checks before publishing or sharing AI-derived results
Syllabus
6 modules · 19 lessons · about 14 hours
1
Module 1: Structure Prediction After AlphaFold
How modern models predict proteins and complexes with ligands, nucleic acids and modifications, and how to judge the output.4 lessons
From AlphaFold 2 to AlphaFold 3: Diffusion, Complexes and the AlphaFold Server
Open Alternatives: Boltz-2, Chai-1 and OpenFold3 Compared
Reading Confidence: pLDDT, PAE, ipTM and Their Blind Spots
Running Predictions at Scale: MSAs, Templates and GPU Memory
2
Module 2: Generative Protein Design
Designing binders, enzymes and scaffolds with diffusion and inverse-folding models.3 lessons
Backbone Generation With RFdiffusion3
Sequence Design With ProteinMPNN and Protein Language Models Such as ESM
Design-Filter-Test: In-Silico Screening, BoltzGen and What the Wet Lab Must Confirm
3
Module 3: Genomics and Variant Interpretation
AI from raw reads to the functional effect of a variant.3 lessons
Deep-Learning Variant Calling: DeepVariant and GPU-Accelerated Parabricks
Variant-Effect Prediction: AlphaMissense and AlphaGenome
DNA and Single-Cell Foundation Models: Evo 2, scGPT and Geneformer
4
Module 4: Biological and Biomedical Imaging
Segmentation, quantification and reconstruction with AI.4 lessons
Cell and Nucleus Segmentation With Cellpose-SAM and micro-SAM in napari
Fine-Tuning on Your Own Images and Measuring Segmentation Quality
MONAI for Medical and Preclinical Imaging Pipelines
AI in Cryo-EM: Particle Picking and Automated Model Building
5
Module 5: Infrastructure, Reproducibility and Licensing
Running these models responsibly on shared compute and publishing results others can trust.3 lessons
GPUs, Containers and Slurm: Packaging Bio-AI Models for an HPC Cluster
Model Licenses and Weights Terms: Academic, Non-Commercial and Open
Reproducibility: Versioning Models, Seeds, Databases and Reporting Methods
6
Module 6: Dual Use and Biosecurity
The responsibilities that come with powerful biological design tools.2 lessons
Dual-Use Risk in Protein and Genome Design Models
Nucleic-Acid Synthesis Screening, Institutional Review and Responsible Release
Exam and certificate
The exam unlocks once you have completed every lesson. It draws 20 questions from a larger bank, spread across the modules, with the answer options shuffled for each attempt. You have 45 minutes and need 80% to pass. After a failed attempt you can try again after 24 hours, up to 3 attempts in total. Feedback is given per module, never as the answer key.
Passing earns a certificate of completion and assessment from the Wirewalk Institute of AI with a unique ID that anyone can check at institute.wirewalk.com/verify. It is not an accredited qualification or a vendor certification. How certification works.
The Wirewalk Institute of AI is independent. This course teaches publicly documented products and is not affiliated with or endorsed by Google DeepMind, Isomorphic Labs, NVIDIA, EvolutionaryScale, Arc Institute, Chai Discovery, Meta, or Hugging Face. Product and company names are trademarks of their respective owners.
A free introduction built on eight healthcare jobs: auditing an ambient-scribe note, answering a portal message from an AI draft, acting on a sepsis alert, reading a triage-flagged CT worklist, reviewing a prior-authorization request, checking AI-suggested codes, and triaging drug and device safety reports. Each lesson shows where AI helps, where it must not decide, and the rule that governs that exact task.
Use a synthetic US care-team case to check AI drafts against source records, preserve nursing judgment and coordinate accountable follow-up. Build a practical evidence packet without patient data or a model account.
Evaluate clinical AI through a fictional US health-system case: procurement, source audits, silent validation, monitoring, privacy and integration. Produce a bounded evidence portfolio using synthetic materials and branching decisions.