Large scale active-learning-guided exploration for in vitro protein production optimization.
Ontology highlight
ABSTRACT:
SUBMITTER: Borkowski O
PROVIDER: S-EPMC7170859 | biostudies-literature | 2020 Apr
REPOSITORIES: biostudies-literature
Publications
Large scale active-learning-guided exploration for in vitro protein production optimization.
Borkowski Olivier O Koch Mathilde M Zettor Agnès A Pandi Amir A Batista Angelo Cardoso AC Soudier Paul P Faulon Jean-Loup JL
Nature communications 20200420 1
Lysate-based cell-free systems have become a major platform to study gene expression but batch-to-batch variation makes protein production difficult to predict. Here we describe an active learning approach to explore a combinatorial space of ~4,000,000 cell-free buffer compositions, maximizing protein production and identifying critical parameters involved in cell-free productivity. We also provide a one-step-method to achieve high quality predictions for protein production using minimal experim ...[more]
Similar Datasets
Project description:Proteins fold to their lowest free-energy structures, and hence the most straightforward way to increase the accuracy of a partially incorrect protein structure model is to search for the lowest-energy nearby structure. This direct approach has met with little success for two reasons: first, energy function inaccuracies can lead to false energy minima, resulting in model degradation rather than improvement; and second, even with an accurate energy function, the search problem is formidable because the energy only drops considerably in the immediate vicinity of the global minimum, and there are a very large number of degrees of freedom. Here we describe a large-scale energy optimization-based refinement method that incorporates advances in both search and energy function accuracy that can substantially improve the accuracy of low-resolution homology models. The method refined low-resolution homology models into correct folds for 50 of 84 diverse protein families and generated improved models in recent blind structure prediction experiments. Analyses of the basis for these improvements reveal contributions from both the improvements in conformational sampling techniques and the energy function.
| S-EPMC5866580 | biostudies-literature
Project description:BackgroundStemphylium blight incited by Stemphylium botryosum poses a significant threat to lentil crops worldwide, inducing severe defoliation and causing substantial yield losses in susceptible varieties under favorable conditions. While some moderate levels of resistance have been identified within lentil germplasm, a low number of resistant cultivars are available to farmers. Adding to the common constraints of resistance breeding, a notable challenge is generating a sufficient number of spores for large-scale screenings, which are essential for pinpointing additional sources of resistance for integration into breeding programs. Therefore, there is a pressing need to improve existing screening methods and tailor them for large-scale material selection. In pursuit of this objective, a protocol for the efficient production of fungal material has been adapted.ResultsOptimization of fungal material production was successfully achieved by comparing the use of fungal mycelia and spores. Spore production was found to be optimal when produced on solid V8-PDA(hi) medium, while liquid Richard's medium was identified as superior for mycelium yield. Furthermore, a refined screening method was developed by evaluating the resistance of six lentil accessions to stemphylium blight. This assessment included the use of either fungal mycelia (at densities ranging from 1 to 5 g L- 1) or spores (with densities ranging from 5 × 104 to 2 × 105 conidia mL- 1) under three different relative humidity levels (from 50 to 100%). Both humidity levels and inoculum dose significantly influenced the final disease rating (DR) and the relative Area Under the Disease Progress Curve (rAUDPC). Differences among genotypes in final symptom severity (DR) became more pronounced after inoculation with inoculum densities of 5 g L- 1 of mycelium or of 105 and 2 × 105 conidia mL- 1 of spore under 100% relative humidity. Given the challenges associated with the large-scale production of S. botryosum spores, inoculations with 5 g L- 1 of mycelium is highly recommended as a practical alternative for conducting mass-scale screenings.ConclusionsThe findings from this study underscore the critical importance of maintaining high level of humidity during inoculation and disease progression development for accurately assessing resistance to stemphylium blight. The optimization of mycelial production for suspension inoculation emerges as a more reliable and efficient approach for conducting large-scale screening to assess germplasm resistance against stemphylium blight in lentil crops.
| S-EPMC10993456 | biostudies-literature
Project description:Ab initio protein docking represents a major challenge for optimizing a noisy and costly "black box"-like function in a high-dimensional space. Despite progress in this field, there is a lack of rigorous uncertainty quantification (UQ). To fill the gap, we introduce a novel algorithm, Bayesian active learning (BAL), for optimization and UQ of such black-box functions with applications to flexible protein docking. BAL directly models the posterior distribution of the global optimum (i.e., native structures) with active sampling and posterior estimation iteratively feeding each other. Furthermore, it uses complex normal modes to span a homogeneous, Euclidean conformation space suitable for high-dimensional optimization and constructs funnel-like energy models for quality estimation of encounter complexes. Over a protein-docking benchmark set and a CAPRI set including homology docking, we establish that BAL significantly improves against starting points from rigid docking and refinements by particle swarm optimization, providing a top-3 near-native prediction for one third targets. Quality assessment empowered with UQ leads to tight quality intervals with half range around 25% of the actual interface root-mean-square deviation and confidence level at 85%. BAL's estimated probability of a prediction being near-native achieves binary classification AUROC at 0.93 and area under the precision recall curve over 0.60 (compared to 0.50 and 0.14, respectively, by chance), which also improves ranking predictions. This study represents the first UQ solution for protein docking, with rigorous theoretical frameworks and comprehensive empirical assessments.
| S-EPMC7429362 | biostudies-literature
Project description:Recently, the wafers used in solar cells have been increasing in size, leading to larger module sizes and weights. The increased weight can cause deflection of photovoltaic (PV) module, which may lead to decreased cell efficiency. In this study, we developed a deep neural network (DNN)-based finite element (FE) surrogate model to obtain the optimal frame design factors that can improve deflection in large-scale bifacial PV module. Initially, an FE model was constructed for large-scale bifacial PV module. Based on this, the FE surrogate model was trained using 243 FEA datasets generated within the proposed range of factors. Furthermore, it was improved through Bayesian optimization and k-fold validation. As a result, the final loss value was 3.743×10-4 , and the average mean absolute percentage error (MAPE) and coefficient of determination ( R2 ) values for deflection and weight were 0.0017, 0.9972 for the training set, and 0.0020, 0.9962 for the test set, respectively. This indicates that the trained FE surrogate model possesses significant accuracy. After generating 1 million datasets within the range of frame design factors, the trained model was used to obtain predictions. Based on this data, the frame design factors that minimize both deflection and weight were identified as about a = 1.5, b = 13.7, c = 1.5, d = 3.0, e = 4.3. At this point, the deflection was 11.1 mm, and the weight was 3.6 kg. After altering the frame shape with the derived factors, FEA was conducted. The results matched for both deflection and weight, with almost no error. At this point, the weight increased by approximately 12.8% compared to the existing, while the deflection decreased by about 9.6%. Additionally, we analyzed the relationship between deflection and weight for each factor and secured the basis for the derived results. Consequently, our FE surrogate model accurately predicted the FEA results and quickly identified the optimal factors that minimize deflection and weight.
| S-EPMC11199489 | biostudies-literature
Project description:This paper aims to propose a table tennis target detection (TD) method based on deep learning (DL) and multi-scale feature fusion (MFF) to improve the detection accuracy of the ball in table tennis competition, optimize the training process of athletes, and improve the technical level. In this paper, DL technology is used to improve the accuracy of table tennis TD through MFF guidance. Initially, based on the FAST Region-based Convolutional Neural Network (FAST R-CNN), the TD is carried out in the table tennis match. Then, through the method of MFF guidance, different levels of feature information are fused, which improves the accuracy of TD. Through the experimental verification on the test set, it is found that the mean Average Precision (mAP) value of the target detection algorithm (TDA) proposed here reaches 87.3%, which is obviously superior to other TDAs and has higher robustness. The DL TDA combined with the proposed MFF can be applied to various detection fields and can help the application of TD in real life.
| S-EPMC10792085 | biostudies-literature
Project description:Plants adapt to fluctuating light conditions by a process called non-photochemical quenching (NPQ), where membrane protein PsbS plays a crucial role and transforms a change in the pH-gradient across the thylakoid membrane under excess light conditions into a photoprotective state, leading to de-excitation of antenna chlorophylls. The PsbS activation mechanism is elusive and has been proposed to involve a monomerization step and protonation of specific residues. To elucidate its function, it is essential to produce PsbS in large quantities, stabilize PsbS in a membrane-mimicking environment and analyze its pH-dependent conformational structure. We present an approach for large-scale in-vitro production and spectroscopic characterization of PsbS under controlled, non-crystalline conditions. We produced PsbS of the moss Physcomitrella patens in milligram quantities in E. coli, refolded PsbS in several detergent types and analyzed its conformation at neutral and low pH by Dynamic Light Scattering and NMR spectroscopy. Our results reveal that at both pH conditions, PsbS exist as dimers or in apparent monomer-dimer equilibria. Lowering of the pH induces conformational changes, destabilizes the dimer state and shifts the equilibria towards the monomeric form. In vivo, a similar response upon thylakoid lumen acidification may tune PsbS activity in a gradual manner.
| S-EPMC5680255 | biostudies-literature
Project description:We provide the first full account of transcriptional changes during terminal erythroid differentiation for cells grown under conditions of stress erythropoiesis. Samples were generated from PBMC's in three-stage culture system in Cellquin, an IMDB based culture medium optimized for erythroid outgrowth. The first two stages rely on presence of EPO Stem Cell Factor and dexamethasone (mimmicking stress erythropoiesis), followed by a terminal differentiation stage in presence of EPO and Holotransferrin. This culture set up allows for generation of large number of cultured Red Blood Cells that ar functionally and morphologically similar to ex-vivo red blood cells.
2019-11-25 | GSE124363 | GEO
Project description:BackgroundBear bile powder is a valuable medicinal material characterized by high content of tauroursodeoxycholic acid (TUDCA) at a certain ratio to taurochenodeoxycholic acid (TCDCA). We had created an engineered E. coli harboring two-step bidirectional oxidative and reductive enzyme-catalyzing pathway that could rapidly convert TCDCA to TUDCA at a specific percentage in shake flasks.ResultsWe reported here the large-scale production of TUDCA containing products by balancing the bidirectional reactions through optimizing fermentation process of the engineered E. coli in fermenters. The fermentation medium was firstly optimized based on M9 medium using response surface methodology, leading to a glycerol and yeast extract modified M9-GY medium benefits for both cell growth and product conversion efficiency. Then isopropylthio-β-galactoside induction and fed-stock stage was successively optimized. Finally, a special deep-tank static process was developed to promote the conversion from TCDCA to TUDCA. Applying the optimal condition, fermentation was performed by separately supplementing 30 g refined chicken bile powder and 35 g crude chicken bile powder as substrates, resulting in 29.35 ± 2.83 g and 30.78 ± 3.04 g powder products containing 35.85 ± 3.85% and 27.14 ± 4.23% of TUDCA at a ratio of 1.49 ± 0.14 and 1.55 ± 0.19 to TCDCA, respectively, after purification and evaporation of the fermentation broth. The recovery yield was 92.84 ± 4.21% and 91.83 ± 2.56%, respectively.ConclusionThis study provided a practical and environment friendly industrialized process for producing artificial substitute of bear bile powder from cheap and readily available chicken bile powder using engineered E. coli microbial cell factory. It also put forward an interesting deep-tank static process to promote the enzyme-catalyzing reactions toward target compounds in synthetic biology-based fermentation.
| S-EPMC6368744 | biostudies-literature
Project description:Machine learning can infer how protein sequence maps to function without requiring a detailed understanding of the underlying physical or biological mechanisms. It is challenging to apply existing supervised learning frameworks to large-scale experimental data generated by deep mutational scanning (DMS) and related methods. DMS data often contain high-dimensional and correlated sequence variables, experimental sampling error and bias, and the presence of missing data. Notably, most DMS data do not contain examples of negative sequences, making it challenging to directly estimate how sequence affects function. Here, we develop a positive-unlabeled (PU) learning framework to infer sequence-function relationships from large-scale DMS data. Our PU learning method displays excellent predictive performance across ten large-scale sequence-function datasets, representing proteins of different folds, functions, and library types. The estimated parameters pinpoint key residues that dictate protein structure and function. Finally, we apply our statistical sequence-function model to design highly stabilized enzymes.
| S-EPMC7856229 | biostudies-literature
Project description:The mankind relies on the use of antibiotics for a healthy life. The epidemic-like emergence of drug-resistant bacterial strains is increasingly becoming one of the leading causes of morbidity and mortality, which gives rise to design a potential antimicrobial peptide (AMP). Here, we have designed the potential AMP using the extensive dynamics simulation since protein-peptide interactions are linked to large conformational changes. Therefore, we have employed the advanced computational avenue CABS molecular docking method that enabled the flexible peptide-protein molecular docking with a large-scale rearrangement of the protein. Lead AMP was investigated against the wild-type (WT) and mutant-PBP5 (MT-PBP5) proteins (antiresistance property). AMP20 showed strong interactions with wtPBP5 and mtPBP5 and involvement of a large number of elements in interactions determined through an atomic model study. Full flexibility analysis showed the stable interaction of AMP20 with both the wild-type and mutant form of PBP5 with root-mean-square deviation (RMSD) values of ∼4.51 and 4.85 Å, respectively. Moreover, peptide dynamics showed involvement of all residues of AMP20 through contact map analysis, and extensive simulation confirmed the stable interaction of AMP20, with lower values of RMSD, radius of gyration, and root-mean-square fluctuation. This study paves the way for a potential approach to design the AMP with amino acid walking and large-scale conformational rearrangements of amino acids.
| S-EPMC6921640 | biostudies-literature