Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper critiques the integration of machine learning in directed evolution, highlighting a disconnect between the objectives of ML-assisted directed evolution (MLDE) researchers and the broader goals of directed evolution. The author argues that the predominant focus on identifying optimal proteins overlooks practical constraints such as the costs of DNA synthesis, limiting the applicability of current MLDE methods. By reflecting on recent exceptions and advocating for a redefined approach to MLDE objectives, the paper aims to bridge this gap and enhance the practical utility of ML in protein engineering.
Current ML-assisted directed evolution strategies often ignore the high costs of DNA synthesis, limiting their real-world applicability despite advances in model capabilities.
The last five-plus years have seen many protein engineering disciplines transformed by advances in machine learning (ML), but the same cannot be said for directed evolution. Reflecting on a previously co-authored perspective, I discuss why I believe this to be the case, arguing that a disconnect between the goals of machine-learning-assisted directed evolution (MLDE) researchers--"identify an optimal protein"--and the goals of directed evolution more broadly--"identify a sufficient protein given time and resource constraints"--is a principal culprit. As an example, I highlight how nearly all current MLDE methods neglect to account for the cost of DNA synthesis, resulting in strategies that have limited practical applicability regardless of the underlying models'capabilities. I close by discussing recent works that are exceptions to this overarching trend, and emphasize that the last five years of efforts in ML-assisted protein engineering and the prescribed reframe of MLDE objectives need not be mutually exclusive.