Adaptive laboratory evolution (ALE) has emerged as a new approach with which to pursue fundamental biological inquiries and, in particular, new insights into the systemic function of a gene product. Two E. coli knockout strains were constructed: one that blocked the Pentose Phosphate Pathway (gnd KO) and one that decoupled the TCA cycle from electron transport (sdhCDAB KO). Despite major perturbations in central metabolism, minimal growth rate changes were found in the two knockout strains. More surprisingly, many similarities were found in their initial transcriptomic states that could be traced to similarly perturbed metabolites despite the differences in the network location of the gene perturbations and concomitant re-routing of pathway fluxes around these perturbations. However, following ALE, distinct metabolomic and transcriptomic states were realized. These included divergent flux and gene expression profiles in the gnd and sdhCDAB KOs to overcome imbalances in NADPH production and nitrogen/sulfur assimilation, respectively, that were not obvious limitations of growth in the unevolved knockouts. Therefore, this work demonstrates that ALE provides a productive approach to reveal novel insights of gene function at a systems level that cannot be found by observing the fresh knockout alone.
Machine learning has greatly advanced over the past decade, owing to advances in algorithmic innovations, hardware acceleration, and benchmark datasets to train on domains such as computer vision, natural-language processing, and more recently the life sciences. In particular, the subfield of machine learning known as deep learning has found applications in genomics, proteomics, and metabolomics. However, a thorough assessment of how the data preprocessing methods required for the analysis of life science data affect the performance of deep learning is lacking. This work contributes to filling that gap by assessing the impact of commonly used as well as newly developed methods employed in data preprocessing workflows for metabolomics that span from raw data to processed data. The results from these analyses are summarized into a set of best practices that can be used by researchers as a starting point for downstream classification and reconstruction tasks using deep learning.
19 May 2023 in npj Systems Biology and Applications
Multi-omics datasets are becoming of key importance to drive discovery in fundamental research as much as generating knowledge for applied biotechnology. However, the construction of such large datasets is usually time-consuming and expensive. Automation might enable to overcome these issues by streamlining workflows from sample generation to data analysis. Here, we describe the construction of a complex workflow for the generation of high-throughput microbial multi-omics datasets. The workflow comprises a custom-built platform for automated cultivation and sampling of microbes, sample preparation protocols, analytical methods for sample analysis and automated scripts for raw data processing. We demonstrate possibilities and limitations of such workflow in generating data for three biotechnologically relevant model organisms, namely Escherichia coli, Saccharomyces cerevisiae, and Pseudomonas putida.
Individuals with long standing diabetes duration can experience damage to small microvascular blood vessels leading to diabetes complications (DCs) and increased mortality. Precision diagnostic tailors a diagnosis to an individual by using biomedical information. Blood small molecule profiling coupled with machine learning (ML) can facilitate the goals of precision diagnostics, including earlier diagnosis and individualized risk scoring. Using data in a cohort of 537 adults with type 1 diabetes (T1D) we predicted five-year progression to DCs. Prediction models were computed first with clinical risk factors at baseline and then with clinical risk factors and blood-derived molecular data at baseline. Progression of diabetic kidney disease and diabetic retinopathy were predicted in two complication-specific models. The model predicts the progression to diabetic kidney disease with accuracy: 0.96 ± 0.25 and 0.96 ± 0.06 area under curve, AUC, with clinical measurements and with small molecule predictors respectively and highlighted main predictors to be albuminuria, glomerular filtration rate, retinopathy status at baseline, sugar derivatives and ketones. For diabetic retinopathy, AUC 0.75 ± 0.14 and 0.79 ± 0.16 with clinical measurements and with small molecule predictors respectively and highlighted key predictors, albuminuria, glomerular filtration rate and retinopathy status at baseline. Individual risk scores were built to visualize results. With further validation ML tools could facilitate the implementation of precision diagnosis in the clinic. It is envisaged that patients could be screened for complications, before these occur, thus preserving healthy life-years for persons with diabetes. This study has been financially supported by Novo Nordisk Foundation grant NNF14OC0013659.
A grand challenge of analytical chemistry is the identification of unknown molecules based on tandem mass spectrometry (MS/MS) spectra. Current metabolite annotation approaches are often manual or partially automated, and commonly rely on a spectral database to search from or train machine learning classifiers on. Unfortunately, spectral databases are often instrument specific and incomplete due to the limited availability of compound standards or a molecular database, which limits the ability of methods utilizing them to predict novel molecule structures. We describe a generative modeling approach that can leverage the vast amount of unpaired and/or unlabeled molecule structures and MS/MS spectra to learn general rules for synthesizing molecule structures and MS/MS spectra. The approach is based on recent work using semi-supervised deep variational autoencoders to learn joint latent representations of multiple and complex modalities. We show that adding molecule structures with no spectra to the training set improves the prediction quality on spectra from a structure disjoint dataset of new molecules, which is not possible using bi-modal supervised approaches. The described methodology provides a demonstration and framework for how recent advances in semi-supervised machine learning can be applied to overcome bottlenecks in missing annotations and noisy data to tackle unaddressed problems in the life sciences where large volumes of data are available.
A biochemical pathway consists of a series of interconnected biochemical reactions to accomplish specific life activities. The participating reactants and resultant products of a pathway, including gene fragments, proteins, and small molecules, coalesce to form a complex reaction network. Biochemical pathways play a critical role in the biochemical domain as they can reveal the flow of biochemical reactions in living organisms, making them essential for understanding life processes. Existing studies of biochemical pathway networks are mainly based on experimentation and pathway database analysis methods, which are plagued by substantial cost constraints. Inspired by the success of representation learning approaches in biomedicine, we develop the biochemical pathway prediction (BPP) platform, which is an automatic BPP platform to identify potential links or attributes within biochemical pathway networks. Our BPP platform incorporates a variety of representation learning models, including the latest hypergraph neural networks technology to model biochemical reactions in pathways. In particular, BPP contains the latest biochemical pathway-based datasets and enables the prediction of potential participants or products of biochemical reactions in biochemical pathways. Additionally, BPP is equipped with an SHAP explainer to explain the predicted results and to calculate the contributions of each participating element. We conduct extensive experiments on our collected biochemical pathway dataset to benchmark the effectiveness of all models available on BPP. Furthermore, our detailed case studies based on the chronological pattern of our dataset demonstrate the effectiveness of our platform. Our BPP web portal, source code and datasets are freely accessible at https://github.com/Glasgow-AI4BioMed/BPP.