Liebal, U. W. | Search bioRxiv

bioRxiv2021

Biotechnology data analysis training with Jupyter Notebooks

Biotechnology has experienced innovations in analytics and data processing. As the volume of data and its complexity grows, new computational procedures for extracting information are developed. However, the rate of change outpaces the adaptation of biotechnology curricula, necessitating new teaching methodologies to equip biotechnologists with data analysis abilities. To simulate experimental data, we created a virtual organism simulator (silvio) by combining diverse cellular and sub-cellular microbial models. silvio was utilized to construct a computer-based instructional workflow with important steps during strain characterization and recombinant protein expression. The instructional workflow is provided as a Jupyter Notebook with comprehensive explanatory text of biotechnological facts and experiment simulations using silvio tools. The students conduct data analysis in Python or Excel. This instructional workflow was separately implemented in two distance courses for Masters students in biology and biotechnology. The concept of using virtual organism simulations that generate coherent results across different experiments can be used to construct consistent and motivating case studies for biotechnological data literacy.

scientific communication and education↗

bioRxiv2020

Exp2Ipynb: A general machine-learning workflow for the analysis of promoter libraries

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. AvailabilityFreely available on GitHub: https://qithub.com/iAMB-RWTH-Aachen/Exp2Ipynb and licensed under the terms of GPLv3. Contactulf.liebal@rwth-aachen.de Supplementary informationSupplementary data available in the Git folder.

bioinformatics↗