Artificial neural networks for classification in metabolomic studies of whole cells using 1H nuclear magnetic resonance.
Ontology highlight
ABSTRACT:
SUBMITTER: Brougham DF
PROVIDER: S-EPMC2945645 | biostudies-literature | 2011
REPOSITORIES: biostudies-literature
Publications
Artificial neural networks for classification in metabolomic studies of whole cells using 1H nuclear magnetic resonance.
Brougham D F DF Ivanova G G Gottschalk M M Collins D M DM Eustace A J AJ O'Connor R R Havel J J
Journal of biomedicine & biotechnology 20100915
We report the successful classification, by artificial neural networks (ANNs), of (1)H NMR spectroscopic data recorded on whole-cell culture samples of four different lung carcinoma cell lines, which display different drug resistance patterns. The robustness of the approach was demonstrated by its ability to classify the cell line correctly in 100% of cases, despite the demonstrated presence of operator-induced sources of variation, and irrespective of which spectra are used for training and for ...[more]
Similar Datasets
Project description:PURPOSE:To investigate the feasibility of using artificial neural networks to estimate stiffness from MR elastography (MRE) data. METHODS:Artificial neural networks were fit using model-based training patterns to estimate stiffness from images of displacement using a patch size of ?1?cm in each dimension. These neural network inversions (NNIs) were then evaluated in a set of simulation experiments designed to investigate the effects of wave interference and noise on NNI accuracy. NNI was also tested in vivo, comparing NNI results against currently used methods. RESULTS:In 4 simulation experiments, NNI performed as well or better than direct inversion (DI) for predicting the known stiffness of the data. Summary NNI results were also shown to be significantly correlated with DI results in the liver (R2 ?=?0.974) and in the brain (R2 ?=?0.915), and also correlated with established biological effects including fibrosis stage in the liver and age in the brain. Finally, repeatability error was lower in the brain using NNI compared to DI, and voxel-wise modeling using NNI stiffness maps detected larger effects than using DI maps with similar levels of smoothing. CONCLUSION:Artificial neural networks represent a new approach to inversion of MRE data. Summary results from NNI and DI are highly correlated and both are capable of detecting biologically relevant signals. Preliminary evidence suggests that NNI stiffness estimates may be more resistant to noise than an algebraic DI approach. Taken together, these results merit future investigation into NNIs to improve the estimation of stiffness in small regions. Magn Reson Med 80:351-360, 2018. © 2017 International Society for Magnetic Resonance in Medicine.
| S-EPMC5876084 | biostudies-literature
Project description:1H NMR spectroscopy and chemometric analysis were used to characterize rat urine obtained after chronic exposure to either tributyl phosphate (TBP) or triphenyl phosphate (TPP). In this study, the daily dose exposure was 1.5 mg/kg body weight for TBP, or 2.0 mg/kg body weight for TPP, administered over a 15-week period. Orthogonal signal correction (OSC) -filtered partial least square discriminant analysis (OSC-PLSDA) was used to predict and classify exposure to these organophosphates. During the development of the model, the classification error was evaluated as a function of the number of latent variables. NMR spectral regions and corresponding metabolites important for determination of exposure type were identified using variable importance in projection (VIP) coefficients obtained from the OSC-PLSDA analysis. As expected, the model for classification of chronic (1.5-2.0 mg/kg body weight daily) TBP or TPP exposure was not as strong as the previously reported model developed for identifying acute (15-20 mg/kg body weight) exposure. The set of majorly impacted metabolites identified for chronic TBP or TPP exposure was slightly different than those metabolites previously identified for acute exposure. These metabolites were then mapped to different metabolite pathways and ranked, allowing the metabolic response to chronic organophosphate exposure to be addressed.
| S-EPMC3901221 | biostudies-literature
Project description:PurposeTo test the hypothesis that removing the assumption of material homogeneity will improve the spatial accuracy of stiffness estimates made by Magnetic Resonance Elastography (MRE).MethodsAn artificial neural network was trained using synthetic wave data computed using a coupled harmonic oscillator model. Material properties were allowed to vary in a piecewise smooth pattern. This neural network inversion (Inhomogeneous Learned Inversion (ILI)) was compared against a previous homogeneous neural network inversion (Homogeneous Learned Inversion (HLI)) and conventional direct inversion (DI) in simulation, phantom, and in-vivo experiments.ResultsIn simulation experiments, ILI was more accurate than HLI and DI in predicting the stiffness of an inclusion in noise-free, low-noise, and high-noise data. In the phantom experiment, ILI delineated inclusions ≤ 2.25 cm in diameter more clearly than HLI and DI, and provided a higher contrast-to-noise ratio for all inclusions. In a series of stiff brain tumors, ILI shows sharper stiffness transitions at the edges of tumors than the other inversions evaluated.ConclusionILI is an artificial neural network based framework for MRE inversion that does not assume homogeneity in material stiffness. Preliminary results suggest that it provides more accurate stiffness estimates and better contrast in small inclusions and at large stiffness gradients than existing algorithms that assume local homogeneity. These results support the need for continued exploration of learning-based approaches to MRE inversion, particularly for applications where high resolution is required.
| S-EPMC7583310 | biostudies-literature
Project description:The complete assignment of the proton chemical shifts obtained by nuclear magnetic resonance (NMR) spectroscopy of de-O-acetylated glucuronoxylomannans (GXMs) from Cryptococcus neoformans permitted the high-resolution determination of the total structure of any GXM. Six structural motifs based on an alpha-(1-->3)-mannotriose substituted with variable quantities of 2-O-beta- and 4-O-beta-xylopyranosyl and 2-O-beta-glucopyranosyluronic acid were identified. The chemical shifts of only the anomeric protons of the mannosyl residues served as structure reporter groups (SRG) for the identification and quantitation of the six triads present in any GXM. The assigned protons for the mannosyl residues resonated at clearly distinguishable positions in the spectrum and supplied all the information essential for the assignment of the complete GXM structure. This technique for assigning structure is referred to as the SRG concept. The SRG concept was used to analyze the distribution of the six mannosyl triads of GXMs obtained from 106 isolates of C. neoformans. The six mannosyl triads occurred singularly or in combination with one or more of the other triads. The identification and quantitation of the SRG were simplified by using a computer-simulated artificial neural network (ANN) to automatically analyze the SRG region of the one-dimensional proton NMR spectra. The occurrence and relative distribution of the six mannosyl triads were used to chemotype C. neoformans on the basis of subtle variations in GXM structure determined by analysis of the SRG region of the proton NMR spectrum by the ANN. The data for the distribution of the six SRGs from GXMs of 106 isolates of C. neoformans yielded eight chemotypes, Chem1 through Chem8.
| S-EPMC121351 | biostudies-literature
Project description:Brain tumors are a diverse group of neoplasms that are challenging to detect and classify due to their varying characteristics. Deep learning techniques have proven to be effective in tumor classification. However, there is a lack of studies that compare these techniques using a common methodology. This work aims to analyze the performance of convolutional neural networks in the classification of brain tumors. We propose a network consisting of a few convolutional layers, batch normalization, and max-pooling. Then, we explore recent deep architectures, such as VGG, ResNet, EfficientNet, or ConvNeXt. The study relies on two magnetic resonance imaging datasets with over 3000 images of three types of tumors -gliomas, meningiomas, and pituitary tumors-, as well as images without tumors. We determine the optimal hyperparameters of the networks using the training and validation sets. The training and test sets are used to assess the performance of the models from different perspectives, including training from scratch, data augmentation, transfer learning, and fine-tuning. The experiments are performed using the TensorFlow and Keras libraries in Python. We compare the accuracy of the models and analyze their complexity based on the capacity of the networks, their training times, and image throughput. Several networks achieve high accuracy rates on both datasets, with the best model achieving 98.7% accuracy, which is on par with state-of-the-art methods. The average precision for each type of tumor is 94.3% for gliomas, 93.8% for meningiomas, 97.9% for pituitary tumors, and 95.3% for images without tumors. VGG is the largest model with over 171 million parameters, whereas MobileNet and EfficientNetB0 are the smallest ones with 3.2 and 5.9 million parameters, respectively. These two neural networks are also the fastest to train with 23.7 and 25.4 seconds per epoch, respectively. On the other hand, ConvNext is the slowest model with 58.2 seconds per epoch. Our custom model obtained the highest image throughput with 234.37 images per second, followed by MobileNet with 226 images per second. ConvNext obtained the smallest throughput with 97.35 images per second. ResNet, MobileNet, and EfficientNet are the most accurate networks, with MobileNet and EfficientNet demonstrating superior performance in terms of complexity. Most models achieve the best accuracy using transfer learning followed by a fine-tuning step. However, data augmentation does not contribute to increasing the accuracy of the models in general.
| S-EPMC10862681 | biostudies-literature
Project description:Human peripheral blood mononuclear cells (PBMCs) are part of the innate and adaptive immune system, and form a critical interface between both systems. Studying the metabolic profile of PBMC could provide valuable information about the response to pathogens, toxins or cancer, the detection of drug toxicity, in drug discovery and cell replacement therapy. The primary purpose of this study was to develop an improved processing method for PBMCs metabolomic profiling with nuclear magnetic resonance (NMR) spectroscopy. To this end, an experimental design was applied to develop an alternative method to process PBMCs at low concentrations. The design included the isolation of PBMCs from the whole blood of four different volunteers, of whom 27 cell samples were processed by two different techniques for quenching and extraction of metabolites: a traditional one using organic solvents and an alternative one employing a high-intensity ultrasound probe, the latter with a variation that includes the use of deproteinizing filters. Finally, all the samples were characterized by 1H-NMR and the metabolomic profiles were compared by the method. As a result, two new methods for PBMCs processing, called Ultrasound Method (UM) and Ultrasound and Ultrafiltration Method (UUM), are described and compared to the Folch Method (FM), which is the standard protocol for extracting metabolites from cell samples. We found that UM and UUM were superior to FM in terms of sensitivity, processing time, spectrum quality, amount of identifiable, quantifiable metabolites and reproducibility.
| S-EPMC7906414 | biostudies-literature
Project description:Phenotype-based assortative mating is well established in humans, with the potential for further convergence through a shared environment. To assess the correlation within infertile couples of physical, social, and behavioural characteristics and 155 circulating metabolic measures. Cross sectional study at a tertiary medical center of 326 couples undertaking IVF. Serum lipids, lipoprotein subclasses, and low-molecular weight metabolites as quantified by NMR spectroscopy (155 metabolic measures). Multivariable and quantile regression correlations within couples of metabolite profiles. Couples exhibited statistical correlations of varying strength for most physical, social, and behavioural characteristics including age, height, alcohol consumption, education, smoking status, physical activity, family history and ethnicity, with correlation coefficients ranging from 0.22 to 0.73. There was no evidence of within couple associations for BMI and weight, where the correlation coefficients were - 0.03 (95% CI - 0.14, 0.08) and 0.01 (95% CI - 0.10, 0.12), respectively. Within spousal associations of the metabolite measurements were all positive but with weak to modest magnitudes, with the median correlation coefficient across all 155 measures being 0.12 (range 0.01-0.37 and interquartile range 0.10-0.18). With just four having associations stronger than 0.3: docosahexaenoic acid (0.37, 95% CI 0.22, 0.52), omega-3 fatty acids (0.32, 95% CI 0.20, 0.43) histidine (0.32, 95% CI 0.23, 0.41) and pyruvate (0.32, 95% CI 0.22, 0.43). Infertile couples exhibit spousal similarities for a range of demographic and serum metabolite measures, supporting initial assortative mating, with diet-derived metabolites suggesting possible subsequent convergence of their individual metabolic profile.
| S-EPMC8566506 | biostudies-literature
Project description:In the present work, a quantitative 1H Nuclear Magnetic Resonance (qHNMR) was established for purity assessment of six aryltetralin lactone lignans. The validation of the method was carried out, including specificity, selectivity, linearity, accuracy, precision, and robustness. Several experimental parameters were optimized, including relaxation delay (D1), scan numbers (NS), and pulse angle. 1,4-Dinitrobenzene was used as internal standard (IS), and deuterated dimethyl sulfoxide (DMSO-d6) as the NMR solvent. The purities were calculated by the area ratios of H-2,6 from target analytes vs. aromatic protons from IS. Six aryltetralin lactone lignans (deoxypodophyllotoxin, podophyllotoxin, 4-demethylpodophyllotoxin, podophyllotoxin-7'-O-?-d-glucopyranoside, 4-demethylpodophyllotoxin-7'-O-?-d-glucopyranoside, and 6''-acetyl-podophyllotoxin-7'-O-? -d-glucopyranoside) were analyzed. The analytic results of qHNMR were further validated by high performance liquid chromatography (HPLC). Therefore, the qHNMR method was a rapid, accurate, reliable tool for monitoring the purity of aryltetralin lactone lignans.
| S-EPMC6272673 | biostudies-literature
Project description:The chemical properties of metal complexes are strongly dependent on the number and geometrical arrangement of ligands coordinated to the metal center. Existing methods for determining either coordination number or geometry rely on a trade-off between accuracy and computational costs, which hinders their application to the study of large structure data sets. Here, we propose MetalHawk (https://github.com/vrettasm/MetalHawk), a machine learning-based approach to perform simultaneous classification of metal site coordination number and geometry through artificial neural networks (ANNs), which were trained using the Cambridge Structural Database (CSD) and Metal Protein Data Bank (MetalPDB). We demonstrate that the CSD-trained model can be used to classify sites belonging to the most common coordination numbers and geometry classes with balanced accuracy equal to 96.51% for CSD-deposited metal sites. The CSD-trained model was also found to be capable of classifying bioinorganic metal sites from the MetalPDB database, with balanced accuracy equal to 84.29% on the whole PDB data set and to 91.66% on manually reviewed sites in the PDB validation set. Moreover, we report evidence that the output vectors of the CSD-trained model can be considered as a proxy indicator of metal-site distortions, showing that these can be interpreted as a low-dimensional representation of subtle geometrical features present in metal site structures.
| S-EPMC11005052 | biostudies-literature
Project description:Gastric cancer (GC) is one of the deadliest cancers worldwide, and the progression of gastric carcinogenesis (GCG) covers multiple complicated pathological stages. Molecular mechanisms of GCG are still unclear. Here, we undertook NMR-based metabolomic analysis of aqueous metabolites extracted from gastric tissues in an established rat model of GCG. We showed that the metabolic profiles were clearly distinguished among 5 histologically classified groups: control, gastritis, low-grade gastric dysplasia, high-grade gastric dysplasia (HGD), and GC. Furthermore, we carried out metabolic pathway analysis based on identified significant metabolites and revealed significantly disturbed metabolic pathways closely associated with the 4 pathological stages, including oxidation stress, choline phosphorylation, amino acid metabolism, Krebs cycle, and glycolysis. Three metabolic pathways were continually disturbed during the progression of GCG, including taurine and hypotaurine metabolism, glutamine and glutamate metabolism, alanine, aspartate, and glutamate metabolism. Both the Krebs cycle and glycine, serine, and threonine metabolism were profoundly impaired in both the HGD and GC stages, potentially due to abnormal energy supply for tumor cell proliferation and growth. Furthermore, valine, leucine, and isoleucine biosynthesis and glycolysis were significantly disturbed in the GC stage for higher energy requirement of the rapid growth of tumor cells. Additionally, we identified potential gastric tissue biomarkers for metabolically discriminating the 4 pathological stages, which also showed good discriminant capabilities for their serum counterparts. This work sheds light on the molecular mechanisms of GCG and is of benefit to the exploration of potential biomarkers for clinically diagnosing and monitoring the progression of GCG.
| S-EPMC7469815 | biostudies-literature