• Protein Optimization is tricky, as evidenced by our recent protein design competition.

• If you don’t have a large budget, you need to be strategic about how you optimize your proteins. Especially if you want to compete in our next round.

• For this blog post, we surveyed the state of the art (SOTA) of using ML protein optimization. We focused on adaptive, cost constrained, and multi-objective methods and have summarized the best papers for you (some of which yielding as much as 7.5-fold improvement, see below!).

• In the next blogpost, we will dive deeper and give concrete, actionable recommendations based on the survey and the learnings from our competition.

• If you already know the field and just want to see our takes a selection of recent papers skip to “Insights from the literature”, or for the full survey with all details, click here.

• If you want to understand them a bit better and need an intro to the underlying concepts, read on.