Variable deep learning training horizons reveal the temporal complexity of biological systems

TL;DR

This paper introduces a deep learning framework that uses variable-length time-series data to study biological systems and identify key transition points.

Contribution

The novel framework allows variable input sequence lengths to better model temporal complexity in biological systems.

Findings

Performance improves with more in silico data but varies with in vitro data.

Temporal dynamics can reveal biological transition points in complex systems.

Abstract

The increasing quantity of time-series images presents new opportunities for extracting biological insights from data. Here, we introduce a deep learning framework with a variable input sequence length to predict cell and colony morphologies. We apply this framework to
in silico
and
in vitro
microscopy datasets, evaluating the impact of temporal data on performance. We find that while performance increases monotonically with increasing
in silico
training data, performance is varied in the
in vitro
case studies. The varying results reflect the intrinsic challenges stochastic, complex biological systems pose to data-driven modeling, and offer a new method through which we can identify biological transition points using temporal dynamics.

Figures