Insight to Gene Expression From Promoter Libraries With the Machine Learning Workflow Exp2Ipynb
Year: 2021
Type: article
Abstract: Strain engineering in biotechnology modifies metabolic pathways in microorganisms to overproduce target metabolites. To modify metabolic pathway activity in bacteria, gene expression is an effective and easy manipulated process, specifically the promoter sequence recognized by sigma factors. Promoter libraries are generated to scan the expression activity of different promoter sequences and to identify sequence positions that predict activity. To maximize information retrieval, a well-designed experimental setup is required. We present a computational workflow to analyse promoter libraries; by applying this workflow to seven libraries, we aim to identify critical design principles. The workflow is based on a Python Jupyter Notebook and covers the following steps: (i) statistical sequence analysis, (ii) sequence-input to expression-output predictions, (iii) estimator performance evaluation, and (iv) new sequence prediction with defined activity. The workflow can process multiple promoter libraries, across species or reporter proteins, and classify or regress expression activity. The strongest predictions in the sample libraries were achieved when the promoters in the library were recognized by a single sigma factor and a unique reporter system. A tradeoff between sample size and sequence diversity reduces prediction quality, and we present a relationship to estimate the minimum sample size. The workflow guides the user through analysis and machine-learning training, is open source and easily adaptable to include alternative machine-learning strategies and to process sequence libraries from other expression-related problems. The workflow is a contribution to increase insight to the growing application of high-throughput experiments and provides support for efficient strain engineering. Availability Freely available on GitHub: https://qithub.com/iAMB-RWTH-Aachen/Exp2Ipynb and licensed under the terms of GPLv3. Contact [email protected] Supplementary information Supplementary data available in the Git folder.
Source: Frontiers in Bioinformatics
Institutions: RWTH Aachen University, Delft University of Technology
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: gold
Estimated APC paid: $1,900
OpenAlex ID: W3207824013