Learning to predict expression efficacy of vectors in recombinant protein production

Abstract

BackgroundRecombinant protein production is a useful biotechnology to produce a large quantity of highly soluble proteins. Currently, the most widely used production system is to fuse a target protein into different vectors in Escherichia coli (E. coli). However, the production efficacy of different vectors varies for different target proteins. Trial-and-error is still the common practice to find out the efficacy of a vector for a given target protein. Previous studies are limited in that they assumed that proteins would be over-expressed and focused only on the solubility of expressed proteins. In fact, many pairings of vectors and proteins result in no expression.ResultsIn this study, we applied machine learning to train prediction models to predict whether a pairing of vector-protein will express or not express in E. coli. For expressed cases, the models further predict whether the expressed proteins would be soluble. We collected a set of real cases from the clients of our recombinant protein production core facility, where six different vectors were designed and studied. This set of cases is used in both training and evaluation of our models. We evaluate three different models based on the support vector machines (SVM) and their ensembles. Unlike many previous works, these models consider the sequence of the target protein as well as the sequence of the whole fusion vector as the features. We show that a model that classifies a case into one of the three classes (no expression, inclusion body and soluble) outperforms a model that considers the nested structure of the three classes, while a model that can take advantage of the hierarchical structure of the three classes performs slight worse but comparably to the best model. Meanwhile, compared to previous works, we show that the prediction accuracy of our best method still performs the best. Lastly, we briefly present two methods to use the trained model in the design of the recombinant protein production systems to improve the chance of high soluble protein production.ConclusionIn this paper, we show that a machine learning approach to the prediction of the efficacy of a vector for a target protein in a recombinant protein production system is promising and may compliment traditional knowledge-driven study of the efficacy. We will release our program to share with other labs in the public domain when this paper is published.

Prediction of recombinant protein overexpression in Escherichia coli using a machine learning based model (RPOLP)

“…Then they applied bioinformatics analysis on the data to determine the key factors for a gene to yield a soluble expression product. According to Chan et al [3], most previous works only investigated protein solubility related factors, and considered together the genes that formed inclusion fraction, and the genes did not express, as the negative samples. But in this study, they built a model to predict whether a vector-protein pair will be expressed in E. coli.…”

“…Based on our literature review, the proposed features in [3] and [16] studies were chosen to generate sequence features. The selection criteria were relevance and the achieved performance.…”

Prediction of soluble heterologous protein expression levels inEscherichia colifrom sequence-based features and its potential in biopharmaceutical process development

“…Interestingly, bioinformatics studies have already shown that primary sequence characteristics have great impact on protein overexpression in E. coli [28]. To explore this relationship further, some useful bioinformatics prediction tools have been established, such as the Wilkinson-Harrison prediction model [29], multiple linear regression fit model [30], solubility index-based model [13], support vector machine (SVM)based model [31,32], PROSO model [12], SOLpro model [33] and PROSO II [14]. One recent review critically summarized the strengths and limitations of seven proposed protein solubility prediction tools in terms of prediction accuracy, Matthews correlation coefficient and data size of studies [22].…”

Correlation between Protein Primary Structure and Soluble Expression Level of HSA dAb in Escherichia coli

“…Several prediction models have been established (6,9), such as the Harrison prediction model (10), multiple linear regression (MLR) model (11), solubility index-based model (12), support vector machine-based model (13,14), PROSO model (15), SOLpro model (16), cc SOL model (17) and PROSO II model (18). These bioinformatics models can significantly reduce trial and error procedures involved in optimization of expression systems to increase the soluble expression level of heterologous proteins.…”