Biology subjects

Engqvist, M. K.

Publications and source records attributed to Engqvist, M. K..

9 recordsLinked to original sources

bioRxiv2022

The substrate scopes of enzymes: a general prediction model based on machine and deep learning

For a comprehensive understanding of metabolism, it is necessary to know all potential substrates for each enzyme encoded in an organisms genome. However, for most proteins annotated as enzymes, it is unknown which primary and/or secondary reactions they catalyze [1], as experimental characterizations are time-consuming and costly. Machine learning predictions could provide an efficient alternative, but are hampered by a lack of information regarding enzyme non-substrates, as available training data comprises mainly positive examples. Here, we present ESP, a general machine learning model for the prediction of enzyme-substrate pairs, with an accuracy of over 90% on independent and diverse test data. This accuracy was achieved by representing enzymes through a modified transformer model [2] with a trained, task-specific token, and by augmenting the positive training data by randomly sampling small molecules and assigning them as non-substrates. ESP can be applied successfully across widely different enzymes and a broad range of metabolites. It outperforms recently published models designed for individual, well-studied enzyme families, which use much more detailed input data [3, 4]. We implemented a user-friendly web server to predict the substrate scope of arbitrary enzymes, which may support not only basic science, but also the development of pharmaceuticals and bioengineering processes.

bioinformatics↗

bioRxiv2022

Learning deep representations of enzyme thermal adaptation

Temperature is a fundamental environmental factor that shapes the evolution of organisms. Learning thermal determinants of protein sequences in evolution thus has profound significance for basic biology, drug discovery, and protein engineering. Here, we use a dataset of over 3 million enzymes labeled with optimal growth temperatures (OGT) of their source organisms to train a deep neural network model (DeepET). The protein-temperature representations learned by DeepET provide a temperature-related statistical summary of protein sequences and capture structural properties that affect thermal stability. For prediction of enzyme optimal catalytic temperatures and protein melting temperatures via a transfer learning approach, our DeepET model outperforms classical regression models trained on rationally designed features and other recent deep-learning-based representations. DeepET thus holds promise for understanding enzyme thermal adaptation and guiding the engineering of thermostable enzymes.

bioinformatics↗

bioRxiv2021

Deep learning based kcat prediction enables improved enzyme constrained model reconstruction

Enzyme turnover numbers (kcat values) are key parameters to understand cell metabolism, proteome allocation and physiological diversity, but experimentally measured kcat data are sparse and noisy. Here we provide a deep learning approach to predict kcat values for metabolic enzymes in a high-throughput manner with the input of substrate structures and protein sequences. Our approach can capture kcat changes for mutated enzymes and identify amino acid residues with great impact on kcat values. Furthermore, we applied the approach to predict genome scale kcat values for over 300 yeast species, demonstrating that the predicted kcat values are consistent with current evolutional understanding. Additionally, we designed an automatic pipeline using the predicted kcat values to parameterize enzyme-constrained genome scale metabolic models (ecGEMs) facilitated by a Bayesian approach, which outperformed the default ecGEMs in predicting phenotypes and proteomes and enabled to explain phenotype differences among yeast species. The deep learning kcat prediction approach and automatic ecGEM construction pipeline would thus be a valuable tool to uncover the global trend of enzyme kinetics and physiological diversity, and to further elucidate cell metabolism on a large scale.

systems biology↗

bioRxiv2020

Experimental investigation of enzyme functional annotations reveals extensive annotation error

Only a small fraction of genes deposited to databases has been experimentally characterised. The majority of proteins have their function assigned automatically, which can result in erroneous annotations. The reliability of current annotations in public databases is largely unknown; experimental attempts to validate the accuracy of existing annotations are lacking. In this study we performed an overview of functional annotations to the BRENDA enzyme database. We first applied a high-throughput experimental platform to verify functional annotations to an enzyme class of S-2-hydroxyacid oxidases (EC 1.1.3.15). We chose 122 representative sequences of the class and screened them for their predicted function. Based on the experimental results, predicted domain architecture and similarity to previously characterised S-2-hydroxyacid oxidases, we inferred that at least 78% of sequences in the enzyme class are misannotated. We experimentally confirmed four alternative activities among the misannotated sequences and showed that misannotation in the enzyme class increased over time. Finally, we performed a computational analysis of annotations to all enzyme classes in BRENDA database, and showed that nearly 18% of all sequences are annotated to an enzyme class while sharing no similarity to experimentally characterised representatives. We showed that even well-studied enzyme classes of industrial relevance are affected by the problem of functional misannotation.

biochemistry↗

bioRxiv2020

Adaptation of a microfluidic qPCR system for enzyme kinetics studies

Microfluidic platforms offer a drastic increase in throughput while minimizing sample usage and hands-on time, which makes them important tools for large-scale biological studies. A range of such systems have been developed for enzyme activity studies, although their complexity largely hinders their application by a wider scientific community. Here we present adaptation of an easy-to-use commercial microfluidic qPCR system for performing enzyme kinetics studies. We demonstrate functionality of the Fluidigm Biomark HD system (the Fluidigm system) by determining kinetic properties of three oxidases in a resorufin-based fluorescent assay. The results obtained in the microfluidic system proved reproducible and comparable to the ones obtained in a standard microplate-based assay. With a wide range of easy-to-use, off-the-shelf components, the microfluidic system presents itself as a simple and customizable platform for high-throughput enzyme activity studies. Graphical abstract O_FIG_DISPLAY_L [Figure 1] M_FIG_DISPLAY C_FIG_DISPLAY

biochemistry↗

bioRxiv2019

Expanding functional protein sequence space using generative adversarial networks

De novo protein design for catalysis of any desired chemical reaction is a long standing goal in protein engineering, due to the broad spectrum of technological, scientific and medical applications. Currently, mapping protein sequence to protein function is, however, neither computationionally nor experimentally tangible 1,2. Here we developed ProteinGAN, a specialised variant of the generative adversarial network 3 that is able to learn natural protein sequence diversity and enables the generation of functional protein sequences. ProteinGAN learns the evolutionary relationships of protein sequences directly from the complex multidimensional amino acid sequence space and creates new, highly diverse sequence variants with natural-like physical properties. Using malate dehydrogenase as a template enzyme, we show that 24% of the ProteinGAN-generated and experimentally tested sequences are soluble and display wild-type level catalytic activity in the tested conditions in vitro, even in highly mutated (>100 mutations) sequences. ProteinGAN therefore demonstrates the potential of artificial intelligence to rapidly generate highly diverse novel functional proteins within the allowed biological constraints of the sequence space.

synthetic biology↗

bioRxiv2019

Age-dependent loss of mitochondrial DNA integrity in mammalian muscle

Ribonucleotides (rNMPs) incorporated in the nuclear genome are a well-established threat to genome stability and can result in DNA strand breaks when not removed in a timely manner. However, the presence of a certain level of rNMPs is tolerated in mitochondrial DNA (mtDNA), although aberrant mtDNA rNMP content has been identified in disease models. We investigated the effect of incorporated rNMPs on mtDNA stability over the mouse lifespan and found that the mtDNA rNMP content increased during early life. The rNMP content of mtDNA varied greatly across different tissues and was defined by the rNTP/dNTP ratio of the tissue. Accordingly, mtDNA rNMPs were nearly absent in SAMHD1-/- mice that have increased dNTP pools. The near absence of rNMPs did not, however, appreciably affect mtDNA copy number or the levels of mtDNA molecules with deletions or strand breaks in aged animals near the end of their lifespan. The physiological rNMP load therefore does not contribute to the progressive loss of mtDNA quality that occurs as mice age.

molecular biology↗

bioRxiv2019

Machine learning applied to predicting microorganism growth temperatures and enzyme catalytic optima

The optimal growth temperature (OGT) of organisms is an important index to estimate the stability of enzymes encoded in their genomes. However, experimental determination of OGT for microorganisms that cannot be cultivated is difficult. Here, we report on the development of a machine learning model that can accurately predict OGT directly from proteome-wide 2-mer amino acid composition. We make use of this model to predict OGTs for 1,438 microorganisms. In a subsequent step we combine OGT data with amino acid composition of individual enzymes to develop a second machine learning model for prediction of enzyme temperature optima (Topt). The resulting model is far superior to using OGT alone for estimating Topt in a dataset of 2,609 enzymes. Finally, we predict Topt for 6.5 million enzymes, covering 4,447 EC numbers, and make the resulting dataset available for researchers, enabling simple identification of enzymes that are potentially functional at extreme temperatures.

bioinformatics↗

bioRxiv2018

Correlating enzyme annotations with a large set of microbial growth temperatures reveals metabolic adaptations to growth at diverse temperatures

Interpreting genomic data to identify temperature adaptations is challenging due to limited accessibility of growth temperature data. In this work I mine public culture collection websites to obtain growth temperature data for 21,498 organisms. Leveraging this unique dataset I identify 319 enzyme activities that either increase or decrease in abundance with temperature. This is a striking result showing that up to 9% of enzyme activities may represent metabolic changes important for adapting to growth at differing temperatures in microbes. Eight metabolic pathways were statistically enriched for these enzyme activities, further highlighting specific areas of metabolism that may be particularly important for such adaptations. Furthermore, I establish a correlation between 33 domains of unknown function (DUFs) with growth temperature in microbes, four of which (DUF438, DUF1524, DUF1957 and DUF3458_C) were significant in both archaea and bacteria. These DUFs may represent novel, as yet undiscovered, functions relating to temperature adaptation.

bioinformatics↗