Week 3 — Lab Automation

cover image cover image

1. Article / case study: Automation at Adaptyv Bio and protein binder design competitions

A prominent example of the use of automation in biology is the work carried out by Adaptyv Bio, a company specialized in laboratory automation and the integration of artificial intelligence for protein design and validation. In particular, Adaptyv Bio organized international protein design competitions, such as the Protein Binder Competition, in which thousands of computationally generated designs were experimentally tested using fully automated workflows.

In these competitions, participants used AI models to design proteins capable of binding to specific therapeutic targets, such as the EGFR receptor and, in similar events, emerging viral proteins such as those from Nipah virus. Subsequently, the best designs were synthesized, expressed, and characterized using automated robotic pipelines, including cloning, protein expression, and affinity assays through Bio-Layer Interferometry (BLI). The entire experimental process was conducted in high-throughput robotic laboratories, enabling the rapid, reproducible, and standardized evaluation of hundreds of proteins.

This approach demonstrated how the integration of artificial intelligence with robotic automation can dramatically accelerate the design–build–test–learn (DBTL) cycle, reducing costs, human error, and experimental time, while simultaneously generating large volumes of reproducible data to improve predictive models.

2. Automation proposal for the final project

For my final project, I plan to implement an automated flow for the design, expression, and functional evaluation of protein binders targeted at essential bacterial targets, combining artificial intelligence, structural bioinformatics, and experimental automation.

Automated general flow

A. Computational design of binders

  • Use of generative AI models and structural prediction (ProteinMPNN, RFdiffusion, AlphaFold2).
  • Molecular docking assessment and molecular dynamics simulations.
  • Automatic prioritization of candidates with better affinity and stability.

B. Experimental automation with Opentrons

  • Automated cloning of binder genes into expression vectors.
  • Bacterial transformation, expression induction and culture preparation.
  • Automated bacterial growth assays and inhibition measurement.

C. Functional validation

  • Automated reading of OD600.
  • Analysis of growth curves.
  • Statistical comparison between controls and strains expressing binders.