Differentiable biology: using deep learning for biophysics-based and data-driven modeling of molecular mechanisms

Deep learning using neural networks relies on a class of machine-learnable models constructed using 'differentiable programs'. These programs can combine mathematical equations specific to a particular domain of natural science with general-purpose, machine-learnable components trained on experimental data. Such programs are having a growing impact on molecular and cellular biology. In this Perspective, we describe an emerging 'differentiable biology' in which phenomena ranging from the small and specific (for example, one experimental assay) to the broad and complex (for example, protein folding) can be modeled effectively and efficiently, often by exploiting knowledge about basic natural phenomena to overcome the limitations of sparse, incomplete and noisy data. By distilling differentiable biology into a small set of conceptual primitives and illustrative vignettes, we show how it can help to address long-standing challenges in integrating multimodal data from diverse experiments across biological scales. This promises to benefit fields as diverse as biophysics and functional genomics. This Perspective describes advances in computer science that enable the integration of deep learning with traditional knowledge-based modeling in biological sciences, and discusses how such integration might overcome the challenges of modeling sparse, incomplete and noisy experimental data.

Automated summary

The perspective discusses the concept of 'differentiable biology' which uses machine-learnable models based on differentiable programs to effectively model biological phenomena. By exploiting knowledge about basic natural phenomena, it overcomes the challenges of sparse, incomplete, and noisy data, benefiting fields such as biophysics and functional genomics. The integration of deep learning with traditional knowledge-based modeling in biological sciences is described as a way to address the challenges of modeling experimental data.