Knowledge transfer across cell lines using Hybrid Gaussian Process models with entity embedding vectors

TL;DR

This paper introduces a hybrid Gaussian Process model with entity embedding vectors to transfer knowledge across cell lines, enabling more efficient predictions and reducing experimental efforts in biochemical process development.

Contribution

It proposes a novel embedding-based approach for Gaussian Process models to effectively transfer knowledge across different products in biochemical processes.

Findings

Embedding vectors capture interpretable product similarities

The method outperforms traditional one-hot encoding in simulations

Potential to significantly reduce wet-lab experiments

Abstract

To date, a large number of experiments are performed to develop a biochemical process. The generated data is used only once, to take decisions for development. Could we exploit data of already developed processes to make predictions for a novel process, we could significantly reduce the number of experiments needed. Processes for different products exhibit differences in behaviour, typically only a subset behave similar. Therefore, effective learning on multiple product spanning process data requires a sensible representation of the product identity. We propose to represent the product identity (a categorical feature) by embedding vectors that serve as input to a Gaussian Process regression model. We demonstrate how the embedding vectors can be learned from process data and show that they capture an interpretable notion of product similarity. The improvement in performance is compared to traditional one-hot encoding on a simulated cross product learning task. All in all, the proposed method could render possible significant reductions in wet-lab experiments.