Variability analysis of LC-MS experimental factors and their impact on machine learning
Ontology highlight
ABSTRACT:
SUBMITTER:
PROVIDER: S-EPMC10659119 | biostudies-literature | 2023 Jan
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Imazalil (IMZ) is a fungicide used in the cultivation of vegetables, such as cucumbers, in green houses or post-harvest on fruit to avoid spoilage due to fungal growth. Agricultural workers can be occupationally exposed to IMZ and the general public indirectly by the diet. The purpose of this study was to develop and validate an LC-MS-MS method for the analysis of IMZ in human urine. The method used electrospray ionization and selected reaction monitoring in the positive mode. Excellent linearity was observed in the range 0.5-100 ng/mL. The limit of detection of the method was 0.2 ng/mL, and the limit of quantitation 0.8 ng/mL. The method showed good within-run, between-run and between-batch precision, with a coefficient of variation <15%. The method was applied to analyze urine samples obtained from two human volunteers following experimental oral and dermal exposure. The excretion of IMZ seemed to follow a two-compartment model and first-order kinetics. In the oral exposure, the elimination half-life of IMZ in the rapid excretion phase was 2.6 and 1.9 h for the female and the male volunteer, respectively. In the slower excretion phase, it was 7.6 and 13 h, respectively. In the dermal exposure, the excretion seemed to follow a single-compartment model and first-order kinetics. The elimination half-life was 10 and 6.6 h for the female and the male volunteer, respectively. Although the study is limited to two volunteers, some information on basic toxicokinetics and metabolism of IMZ in humans is presented.
| S-EPMC4626574 | biostudies-literature
Project description:Recent mass spectrometry (MS)-based techniques enable deep proteome coverage with relative quantitative analysis, resulting in increased identification of very weak signals accompanied by increased data size of liquid chromatography (LC)-MS/MS spectra. However, the identification of weak signals using an assignment strategy with poorer performance results in imperfect quantification with misidentification of peaks and ratio distortions. Manually annotating a large number of signals within a very large dataset is not a realistic approach. In this study, therefore, we utilized machine learning algorithms to successfully extract a higher number of peptide peaks with high accuracy and precision. Our strategy evaluated each peak identified using six different algorithms; peptide peaks identified by all six algorithms (i.e., unanimously selected) were subsequently assigned as true peaks, which resulted in a reduction in the false-positive rate. Hence, exact and highly quantitative peptide peaks were obtained, providing better performance than obtained applying the conventional criteria or using a single machine learning algorithm.
| S-EPMC8642397 | biostudies-literature
Project description:Fast identification of microbial species in clinical samples is essential to provide an appropriate antibiotherapy to the patient and reduce the prescription of broad-spectrum antimicrobials leading to antibioresistances. MALDI-TOF-MS technology has become a tool of choice for microbial identification but has several drawbacks: it requires a long step of bacterial culture before analysis (≥24 h), has a low specificity and is not quantitative. We developed a new strategy for identifying bacterial species in urine using specific LC-MS/MS peptidic signatures. In the first training step, libraries of peptides are obtained on pure bacterial colonies in DDA mode, their detection in urine is then verified in DIA mode, followed by the use of machine learning classifiers (NaiveBayes, BayesNet and Hoeffding tree) to define a peptidic signature to distinguish each bacterial species from the others. Then, in the second step, this signature is monitored in unknown urine samples using targeted proteomics. This method, allowing bacterial identification in less than 4 h, has been applied to fifteen species representing 84% of all Urinary Tract Infections. More than 31,000 peptides in 190 samples were quantified by DIA and classified by machine learning to determine an 82 peptides signature and build a prediction model. This signature was validated for its use in routine using Parallel Reaction Monitoring on two different instruments. Linearity and reproducibility of the method were demonstrated as well as its accuracy on donor specimens. Within 4h and without bacterial culture, our method was able to predict the predominant bacteria infecting a sample in 97% of cases and 100% above the standard threshold. This work demonstrates the efficiency of our method for the rapid and specific identification of the bacterial species causing UTI and could be extended in the future to other biological specimens and to bacteria having specific virulence or resistance factors.
| S-EPMC6885708 | biostudies-literature
Project description:Proteins extracted from mouse liver were digested by trypsin, then tryptic peptides were analyzed using Q-Exactive.
2022-08-09 | PXD027824 | JPOST Repository
Project description:ObjectiveTo implement and evaluate machine learning (ML) algorithms for the prediction of COVID-19 diagnosis, severity, and fatality and to assess biomarkers potentially associated with these outcomes.Material and methodsSerum (n = 96) and plasma (n = 96) samples from patients with COVID-19 (acute, severe and fatal illness) from two independent hospitals in China were analyzed by LC-MS. Samples from healthy volunteers and from patients with pneumonia caused by other viruses (i.e. negative RT-PCR for COVID-19) were used as controls. Seven different ML-based models were built: PLS-DA, ANNDA, XGBoostDA, SIMCA, SVM, LREG and KNN.ResultsThe PLS-DA model presented the best performance for both datasets, with accuracy rates to predict the diagnosis, severity and fatality of COVID-19 of 93%, 94% and 97%, respectively. Low levels of the metabolites ribothymidine, 4-hydroxyphenylacetoylcarnitine and uridine were associated with COVID-19 positivity, whereas high levels of N-acetyl-glucosamine-1-phosphate, cysteinylglycine, methyl isobutyrate, l-ornithine and 5,6-dihydro-5-methyluracil were significantly related to greater severity and fatality from COVID-19.ConclusionThe PLS-DA model can help to predict SARS-CoV-2 diagnosis, severity and fatality in daily practice. Some biomarkers typically increased in COVID-19 patients' serum or plasma (i.e. ribothymidine, N-acetyl-glucosamine-1-phosphate, l-ornithine, 5,6-dihydro-5-methyluracil) should be further evaluated as prognostic indicators of the disease.
| S-EPMC9123826 | biostudies-literature
Project description:Preprocessing of liquid chromatography-mass spectrometry (LC-MS) raw data facilitates downstream statistical and biological data analyses. In the case of targeted LC-MS data, consistent recognition of chromatographic peaks is a main challenge, in particular, for low abundant signals. Fully automatic preprocessing is faster than manual peak review and does not depend on the individual operator. Here, we present the R package automRm for fully automatic preprocessing of LC-MS data recorded in MRM mode. Using machine learning (ML) for detection of chromatographic peaks and quality control of reported results enables the automatic recognition of complex patterns in raw data. In addition, this approach renders automRm generally applicable to a wide range of analytical methods including hydrophilic interaction liquid chromatography (HILIC), which is known for sample-to-sample variations in peak shape and retention time. We demonstrate the impact of the choice of training data set, of the applied ML algorithm, and of individual peak characteristics on automRm's ability to correctly report chromatographic peaks. Next, we show that automRm can replicate results obtained by manual peak review on published data. Moreover, automRm outperforms alternative software solutions regarding the variation in peak integration among replicate measurements and the number of correctly reported peaks when applied to a HILIC-MS data set. The R package is freely available from gitlab (https://gitlab.gwdg.de/joerg.buescher/automrm).
| S-EPMC9047440 | biostudies-literature
Project description:To design a robust quantitative proteomics study, an understanding of both the inherent heterogeneity of the biological samples being studied as well as the technical variability of the proteomics methods and platform is needed. Additionally, accurately identifying the technical steps associated with the largest variability would provide valuable information for the improvement and design of future processing pipelines. We present an experimental strategy that allows for a detailed examination of the variability of the quantitative LC-MS proteomics measurements. By replicating analyses at different stages of processing, various technical components can be estimated and their individual contribution to technical variability can be dissected. This design can be easily adapted to other quantitative proteomics pipelines. Herein, we applied this methodology to our label-free workflow for the processing of human brain tissue. For this application, the pipeline was divided into four critical components: Tissue dissection and homogenization (extraction), protein denaturation followed by trypsin digestion and SPE cleanup (digestion), short-term run-to-run instrumental response fluctuation (instrumental variance), and long-term drift of the quantitative response of the LC-MS/MS platform over the 2 week period of continuous analysis (instrumental stability). From this analysis, we found the following contributions to variability: extraction (72%) >> instrumental variance (16%) > instrumental stability (8.4%) > digestion (3.1%). Furthermore, the stability of the platform and its suitability for discovery proteomics studies is demonstrated.
| S-EPMC3695475 | biostudies-literature
Project description:The inflammatory system activated by uterine infection is associated with decreased fertility. Diseases can be detected in advance by identifying biomarkers of several uterine diseases. Escherichia coli is one of the most frequent bacteria that is involved in pathogenic processes in dairy goats. The purpose of this study was to investigate the effect of endotoxin on protein expression in goat endometrial epithelial cells. In this study, the LC-MS/MS approach was employed to investigate the proteome profile of goat endometrial epithelial cells. A total of 1180 proteins were identified in the goat Endometrial Epithelial Cells and LPS-treated goat Endometrial Epithelial Cell groups, of which, 313 differentially expressed proteins were accurately screened. The proteomic results were independently verified by WB, TEM and IF techniques, and the same conclusion was obtained. To conclude, this model is suitable for the further study of infertility caused by endometrial damage caused by endotoxin. These findings may provide useful information for the prevention and treatment of endometritis.
| S-EPMC10298677 | biostudies-literature
Project description:This article describes processing of protein samples using 1D SDS gels prior to protease digestion for proteomics workflows that subsequently utilize reversed-phase nanocapillary ultra-high-pressure liquid chromatography (LC) coupled to tandem mass spectrometry (MS/MS). The resulting LC-MS/MS data are used to identify peptides and thereby infer proteins present in samples ranging from simple mixtures to very complex proteomes. Bottom-up proteome studies usually involve quantitative comparisons across several or many samples. For either situation, 1D SDS gels represent a simple, widely available technique that can be used to either fractionate complex proteomes or rapidly clean up low microgram samples with minimal losses. After gel separation and staining/destaining, appropriate gel slices are excised, and in-gel reduction, alkylation, and protease digestion are performed. Digests are then processed for LC-MS/MS analysis. Protocols are described for either sample fractionation with high-throughput processing of many samples or simple cleanup without fractionation. An optional strategy is to conduct in-solution reduction and alkylation prior to running gels, which is advantageous when a large number of samples will be separated into large numbers of fractions. Optimization of trypsin digestion parameters and comparison to in-solution protease digestion are also described. © 2019 by John Wiley & Sons, Inc.
| S-EPMC6653605 | biostudies-literature
Project description:A dataset of liquid chromatography-mass spectrometry measurements of medicinal plant extracts from 74 species was generated and used for training and validating plant species identification algorithms. Various strategies for data handling and feature space extraction were tested. Constrained Tucker decomposition, large-scale (more than 1500 variables) discrete Bayesian Networks and autoencoder based dimensionality reduction coupled with continuous Bayes classifier and logistic regression were optimized to achieve the best accuracy. Even with elimination of all retention time values accuracies of up to 96% and 92% were achieved on validation set for plant species and plant organ identification respectively. Benefits and drawbacks of used algortihms were discussed. Preliminary test showed that developed approaches exhibit tolerance to changes in data created by using different extraction methods and/or equipment. Dataset with more than 2200 chromatograms was published in an open repository.
| S-EPMC6243014 | biostudies-literature