Machine learning to design integral membrane channelrhodopsins for efficient eukaryotic expression and plasma membrane localization
By Claire N. Bedbrook, Kevin Kaichuang Yang, Austin J. Rice, Viviana Gradinaru, Frances H. Arnold. 2017
112 citations recorded by the source corpus.
There is growing interest in studying and engineering integral membrane proteins (MPs) that play key roles in sensing and regulating cellular response to diverse external signals. A MP must be expressed, correctly inserted and folded in a lipid bilayer, and trafficked to the proper cellular location in order to function. The sequence and structural determinants of these processes are complex and highly constrained. Here we describe a predictive, machine-learning approach that captures this complexity to facilitate successful MP engineering and design. Machine learning on carefully-chosen training sequences made by structure-guided SCHEMA recombination has enabled us to accurately predict the rare sequences in a diverse library of channelrhodopsins (ChRs) that express and localize to the plasma membrane of mammalian cells. These light-gated channel proteins of microbial origin are of interest for neuroscience applications, where expression and localization to the plasma membrane is a prerequisite for function. We trained Gaussian process (GP) classification and regression models with expression and localization data from 218 ChR chimeras chosen from a 118,098-variant library designed by SCHEMA recombination of three parent ChRs. We use these GP models to identify ChRs that express and localize well and show that our models can elucidate sequence and structure elements important for these processes. We also used the predictive models to convert a naturally occurring ChR incapable of mammalian localization into one that localizes well.
Subfields: Biomedical Engineering, Cellular and Molecular Neuroscience, Cognitive Neuroscience
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Biomedical Engineering
In Cellular and Molecular Neuroscience
In Cognitive Neuroscience
Citations in the corpus
- Evolutionary context-integrated deep sequence modeling for protein engineering (2020, methodology): “To reduce the experimental effort, machine learning (ML) methods have been proposed to assist directed evolution and successfully engineered multiple important functional proteins [15–20] .”
- Machine-learning-guided directed evolution for protein engineering (2018, methodology): “GPs have been used to predict thermostability 11,32,43 , substrates for enzymatic reactions 44 , fluorescence 45 , membrane localization 12 , and channelrhodopsin photoproperties 13 .”
- Learned protein embeddings for machine learning (2018, methodology): “A protein sequence can be encoded by its physical properties or directly by its amino acids (Alipanahi et al., 2015; Bedbrook et al., 2017b; Chang et al., 2016; Fox et al., 2007; Ofer and Linial, 2015; Romero et al., 2013; Saladi et al., 2018).”
- Toward machine-guided design of proteins (2018, background): “Previous work leveraging machine learning methods to engineer proteins focused on variations of linear regression [12, 13], Gaussian process (GP) regression [14, 15], and other kernel-based methods to model the relationship between sequence and fitness to produce improved proteins.”
- Machine learning in protein engineering (2018, methodology): “…have been used to predict thermostability (Pires et al., 2013; Romero et al., 2013b; Jokinen et al., 2018), enzyme substrates (Mellor et al., 2016), fluorescence (Saito et al., 2018), membrane localization (Bedbrook et al., 2017), and channelrhodopsin photo-properties (Bedbrook et al., 2018).”