Enzyme commission (EC) numbers, which associate a protein sequence with the biochemical reactions it catalyzes, are essential for the accurate understanding of enzyme functions and cellular metabolism. Many ab initio computational approaches were proposed to predict EC numbers for given input protein sequences. However, the prediction performance (accuracy, recall, and precision), usability, and efficiency of existing methods decreased seriously when dealing with recently discovered proteins, thus still having much room to be improved. Here, we report HDMLF, a hierarchical dual-core multitask learning framework for accurately predicting EC numbers based on novel deep learning techniques. HDMLF is composed of an embedding core and a learning core; the embedding core adopts the latest protein language model for protein sequence embedding, and the learning core conducts the EC number prediction. Specifically, HDMLF is designed on the basis of a gated recurrent unit framework to perform EC number prediction in the multi-objective hierarchy, multitasking manner. Additionally, we introduced an attention layer to optimize the EC prediction and employed a greedy strategy to integrate and fine-tune the final model. Comparative analyses against 4 representative methods demonstrate that HDMLF stably delivers the highest performance, which improves accuracy and F1 score by 60% and 40% over the state of the art, respectively. An additional case study of tyrB predicted to compensate for the loss of aspartate aminotransferase aspC, as reported in a previous experimental study, shows that our model can also be used to uncover the enzyme promiscuity. Finally, we established a web platform, namely, ECRECer ( https://ecrecer.biodesign.ac.cn ), using an entirely could-based serverless architecture and provided an offline bundle to improve usability.
Background: The Canadian Cattle Genome Project is a large-scale international project that aims to develop genomics-based tools to enhance the efficiency and sustainability of beef and dairy production. Obtaining DNA sequence information is an important part of achieving this goal as it facilitates efforts to associate specific DNA differences with phenotypic variation. These associations can be used to guide breeding decisions and provide valuable insight into the molecular basis of traits. Findings: We describe a dataset of 379 whole-genome sequences, taken primarily from key historic Bos taurus animals, along with the analyses that were performed to assess data quality. The sequenced animals represent ten populations relevant to beef or dairy production. Animal information (name, breed, population), sequence data metrics (mapping rate, depth, concordance), and sequence repository identifiers (NCBI BioProject and BioSample IDs) are provided to enable others to access and exploit this sequence information. Conclusions: The large number of whole-genome sequences generated as a result of this project will contribute to ongoing work aiming to catalogue the variation that exists in cattle as well as efforts to improve traits through genotype-guided selection. Studies of gene function, population structure, and sequence evolution are also likely to benefit from the availability of this resource.
Flux balance analysis (FBA) is an important method for calculating optimal pathways to produce industrially important chemicals in genome-scale metabolic models (GEMs). However, for biologists, the requirement of coding skills poses a significant obstacle to using FBA for pathway analysis and engineering target identification. Additionally, a time-consuming manual drawing process is often needed to illustrate the mass flow in an FBA-calculated pathway, making it challenging to detect errors or discover interesting metabolic features. To solve this problem, we developed CAVE, a cloud-based platform for the integrated calculation, visualization, examination and correction of metabolic pathways. CAVE can analyze and visualize pathways for over 100 published GEMs or user-uploaded GEMs, allowing for quicker examination and identification of special metabolic features in a particular GEM. Additionally, CAVE offers model modification functions, such as gene/reaction removal or addition, making it easy for users to correct errors found in pathway analysis and obtain more reliable pathways. With a focus on the design and analysis of optimal pathways for biochemicals, CAVE complements existing visualization tools based on manually drawn global maps and can be applied to a broader range of organisms for rational metabolic engineering. CAVE is available at https://cave.biodesign.ac.cn/.
1 September 2024 in Synthetic and Systems Biotechnology
Genome-scale metabolic models (GEMs) have been widely employed to predict microorganism behaviors. However, GEMs only consider stoichiometric constraints, leading to a linear increase in simulated growth and product yields as substrate uptake rates rise. This divergence from experimental measurements prompted the creation of enzyme-constrained models (ecModels) for various species, successfully enhancing chemical production. Building upon studies that allocate macromolecule resources, we developed a Python-based workflow (ECMpy) that constructs an enzyme-constrained model. This involves directly imposing an enzyme amount constraint in GEM and accounting for protein subunit composition in reactions. However, this procedure demands manual collection of enzyme kinetic parameter information and subunit composition details, making it rather user-unfriendly. In this work, we've enhanced the ECMpy toolbox to version 2.0, broadening its scope to automatically generate ecGEMs for a wider array of organisms. ECMpy 2.0 automates the retrieval of enzyme kinetic parameters and employs machine learning for predicting these parameters, which significantly enhances parameter coverage. Additionally, ECMpy 2.0 introduces common analytical and visualization features for ecModels, rendering computational results more user accessible. Furthermore, ECMpy 2.0 seamlessly integrates three published algorithms that exploit ecModels to uncover potential targets for metabolic engineering. ECMpy 2.0 is available at https://github.com/tibbdc/ECMpy or as a pip package (https://pypi.org/project/ECMpy/).
Whole genome sequences (WGS) have proliferated as sequencing technology continues to improve and costs decline. While many WGS of model or domestic organisms have been produced, a growing number of non-model species are also being sequenced. In the absence of a reference, construction of a genome sequence necessitates de novo assembly which may be beyond the ability of many labs due to the large volumes of raw sequence data and extensive bioinformatics required. In contrast, the presence of a reference WGS allows for alignment which is more tractable than assembly. Recent work has highlighted that the reference need not come from the same species, potentially enabling a wide array of species WGS to be constructed using cross-species alignment. Here we report on the creation a draft WGS from a single bighorn sheep (Ovis canadensis) using alignment to the closely related domestic sheep (Ovis aries). Two sequencing libraries on SOLiD platforms yielded over 865 million reads, and combined alignment to the domestic sheep reference resulted in a nearly complete sequence (95% coverage of the reference) at an average of 12x read depth (104 SD). From this we discovered over 15 million variants and annotated them relative to the domestic sheep reference. We then conducted an enrichment analysis of those SNPs showing fixed differences between the reference and sequenced individual and found significant differences in a number of gene ontology (GO) terms, including those associated with reproduction, muscle properties, and bone deposition. Our results demonstrate that cross-species alignment enables the creation of novel WGS for non-model organisms. The bighorn sheep WGS will provide a resource for future resequencing studies or comparative genomics.