Metabolomics Analysis in Acute Paraquat Poisoning Patients Based on UPLC-Q-TOF-MS and Machine Learning Approach

Application and Progress of Machine Learning in Pesticide Hazard and Risk Assessment

Long-term exposure to pesticides is associated with the incidence of cancer. With the exponential increase in the number of new pesticides being synthesized, it becomes more and more important to evaluate the toxicity of pesticides by means of simulated calculations. Based on existing data, machine learning methods can train and model the predictions of the effects of novel pesticides, which have limited available data. Combined with other technologies, this can aid the synthesis of new pesticides with specific active structures, detect pesticide residues, and identify their tolerable exposure levels. This article mainly discusses support vector machines, linear discriminant analysis, decision trees, partial least squares, and algorithms based on feedforward neural networks in machine learning. It is envisaged that this article will provide scientists and users with a better understanding of machine learning and its application prospects in pesticide toxicity assessment.

Integrated untargeted and targeted proteomics to unveil plasma prognostic markers for patients with acute paraquat poisoning: A pilot study

Paraquat (PQ) is a widely used but strongly toxic herbicide, which can induce multiple organ failure. The overall survival rate of the poisoned patients was only 54.4% due to lack of specific antidotes. Besides, the definite pathogenic mechanism of PQ is still not fully understood. In this pilot study, untargeted and targeted proteomics were integrated to explore the expression characteristics of plasma protein in PQ poisoned patients, and identify the differentially expressed proteins between survivors and non-survivors. A total of 494 plasma proteins were detected, and of which 47 were upregulated and 44 were downregulated in PQ poisoned patients compared to healthy controls. Among them, five differential plasma proteins (S100A9, S100A8, MB, ACTB and RAB11FIP3) were further validated by multiple reaction monitoring (MRM)-based targeted proteomic approach, and three of them (S100A9, S100A8 and ACTB) were confirmed to be correlated with PQ poisoning. Meanwhile, 84 dysregulated plasma proteins were identified in non-survivors compared with survivors. Moreover, targeted proteomic and ROC analysis suggested that ACTB had a good performance in predicting the prognosis of PQ poisoned patients. These findings highlighted the value of label-free and mass spectrometry-based proteomics in screening prognostic biomarkers of PQ poisoning and studying the mechanism of PQ toxicity.

Abnormalities of serum lipid metabolism in patients with acute paraquat poisoning caused by ferroptosis

As the mechanism of paraquat (PQ) poisoning is still not fully elucidated, and no specific treatment has been developed in medical practice, the management of PQ poisoning continues to present a medical challenge. In this study, the objective was to investigate the early metabolic changes in serum metabolism and identify the key metabolic pathways involved in patients with PQ poisoning. Quantitative analysis was conducted to determine the relevant metabolites. Additionally, experiments were carried out in both plasma and cell to elucidate the mechanisms underlying metabolic disorder and cell death in PQ poisoning. The study found that polyunsaturated fatty acids (PUFAs) and their metabolites, such as arachidonic acid (AA) and hydroxy eicosatetraenoic acids (HETEs), were significantly increased by non-enzymatic oxidative reaction. Reactive oxygen species (ROS) production increased rapidly at 2 h after PQ poisoning, followed by an increase in PUFAs at 12 h, and intracellular glutathione, cysteine (Cys), and Fe2+ at 24 h. However, at 36 h later, intracellular glutathione and Cys decreased, HETEs increased, and the expression of SLC7A11 and glutathione peroxidase 4 (GPX4) decreased. Ultrastructural examination revealed the absence of mitochondrial cristae. Deferoxamine was found to alleviate lipid oxidation, and increase the viability of PQ toxic cells in the low dose. In conclusion, unsaturated fatty acids metabolism was the key metabolic pathways in PQ poisoning. PQ caused cell death through the induction of ferroptosis. Inhibition of ferroptosis could be a novel strategy for the treatment of PQ poisoning.

The value of machine learning for prognosis prediction of diphenhydramine exposure: National analysis of 50,000 patients in the United States

Background

Diphenhydramine (DPH) is an antihistamine medication that in overdose can result in anticholinergic symptoms and serious complications, including arrhythmia and coma. We aimed to compare the value of various machine learning (ML) models, including light gradient boosting machine (LGBM), logistic regression (LR), and random forest (RF), in the outcome prediction of DPH poisoning.

Materials and Methods

We used the National Poison Data System database and included all of the human exposures of DPH from January 01, 2017 to December 31, 2017, and excluded those cases with missing information, duplicated cases, and those who reported co-ingestion. Data were split into training and test datasets, and three ML models were compared. We developed confusion matrices for each, and standard performance metrics were calculated.

Results

Our study population included 53,761 patients with DPH exposure. The most common reasons for exposure, outcome, chronicity of exposure, and formulation were captured. Our results showed that the average precision-recall area under the curve (AUC) of 0.84. LGBM and RF had the highest performance (average AUC of 0.91), followed by LR (average AUC of 0.90). The specificity of the models was 87.0% in the testing groups. The precision of models was 75.0%. Recall (sensitivity) of models ranged between 73% and 75% with an F1 score of 75.0%. The overall accuracy of LGBM, LR, and RF models in the test dataset was 74.8%, 74.0%, and 75.1%, respectively. In total, just 1.1% of patients (mostly those with major outcomes) received physostigmine.

Conclusion

Our study demonstrates the application of ML in the prediction of DPH poisoning.

Metabolomics profile of plasma in acute diquat-poisoned patients using gas chromatography-mass spectrometry

Diquat (DQ) has been confirmed to be toxic to humans and responsible for severe health impairment. While to date, very little is known about the toxicological mechanisms of DQ. Thus, investigations to discover the toxic targets and potential biomarkers of DQ poisoning are urgently needed. In this study, a metabolic profiling analysis was conducted to reveal the changes of metabolites of plasma and find out the potential biomarkers of DQ intoxication by GC-MS. First, multivariate statistical analysis demonstrated that acute DQ poisoning can lead to metabolomic changes in human plasma. Then, metabolomics studies showed that 31 of the identified metabolites were significantly altered by DQ. Pathway analysis indicated that three primarily metabolic pathways including phenylalanine, tyrosine and tryptophan biosynthesis, taurine and hypotaurine metabolism, and phenylalanine metabolism were affected by DQ, resulting in the perturbations of phenylalanine, tyrosine, taurine, and cysteine. Finally, the results of receiver operating characteristic analysis showed the above four metabolites could be used as reliable tools for the diagnosis and severity assessments of DQ intoxication. These data provided the theoretical basis for basic research to understand the potential mechanisms of DQ poisoning, and also identified the desirable biomarkers with great potential for clinical applications.

Utility of artificial intelligence to identify antihyperglycemic agents poisoning in the USA: introducing a practical web application using National Poison Data System (NPDS)

Clinical effects of antihyperglycemic agents poisoning may overlap each other. So, distinguishing exposure to these pharmaceutical drugs may take work. This study examined the application of machine learning techniques in identifying antihyperglycemic agent exposure using the national poisoning database in the USA. In this study, the data of single exposure due to Biguanides and Sulfonylureas (n=6183) was requested from the National Poison Data System (NPDS) for 2014-2018. We have tried five machine learning models (random forest classifier, k-nearest neighbors, Xgboost classifier, logistic regression, neural network Keras). For the multiclass classification modeling, we have divided the dataset into two parts: train (75%) and test (25%). The performance metrics used were accuracy, specificity, precision, recall, and F1-score. The algorithms used to get the classification results of different models to diagnose antihyperglycemic agents were very accurate. The accuracy of our model in determining these two antihyperglycemic agents was 91-93%. The precision-recall curve showed average precision of 0.91, 0.97, 0.97, and 0.98 for k-nearest neighbors, logistic regression, random forest, and XGB, respectively. The logistic regression, random forest, and XGB had the highest AUC (AUC=0.97) among both biguanides and sulfonylureas groups. The negative predictive values (NPV) for all the models were between 89 and 93%. We introduced a practical web application to help physicians distinguish between these agents. Despite variations in accuracy among the different types of algorithms used, all of them could accurately determine the specific exposure to biguanides and sulfonylureas retrospectively. Machine learning can distinguish antihyperglycemic agents, which may be useful for physicians without any background in medical toxicology. Besides, Our suggested ML-based Web application might help physicians in their diagnosis.

Paraquat and Diquat: Recent Updates on Their Pretreatment and Analysis Methods since 2010 in Biological Samples

Paraquat (PQ) and diquat (DQ) are quaternary ammonium herbicides which have been used worldwide for controlling the growth of weeds on land and in water. However, PQ and DQ are well known to be toxic. PQ is especially toxic to humans. Moreover, there is no specific antidote for PQ poisoning. The main treatment for PQ poisoning is hemoperfusion to reduce the PQ concentration in blood. Therefore, it is essential to be able to detect PQ and DQ concentrations in biological samples. This critical review summarizes the articles published from 2010 to 2022 and can help researchers to understand the development of the sample treatment and analytical methods for the determination of PQ and DQ in various types of biological samples. The sample preparation includes liquid-liquid extraction, solid-phase extraction based on different novel materials, microextration methods, and other methods. Analytical methods for quantifying PQ and DQ, such as different chromatography and spectroscopy methods, electrochemical methods, and immunological methods, are illustrated and compared. We focus on the latest advances in PQ and DQ treatment and the application of new technologies for these analyses. In our opinion, tandem mass spectrometry is a good choice for the determination of PQ and DQ, due to its high sensitivity, high selectivity, and high accuracy. As far as we are concerned, the best LOD of 4 pg/mL for PQ in serum can be obtained.

Deep learning neural network derivation and testing to distinguish acute poisonings

INTRODUCTION

Acute poisoning is a significant global health burden, and the causative agent is often unclear. The primary aim of this pilot study was to develop a deep learning algorithm that predicts the most probable agent a poisoned patient was exposed to from a pre-specified list of drugs.

RESEARCH DESIGN & METHODS

Data were queried from the National Poison Data System (NPDS) from 2014 through 2018 for eight single-agent poisonings (acetaminophen, diphenhydramine, aspirin, calcium channel blockers, sulfonylureas, benzodiazepines, bupropion, and lithium). Two Deep Neural Networks (PyTorch and Keras) designed for multi-class classification tasks were applied.

RESULTS

There were 201,031 single-agent poisonings included in the analysis. For distinguishing among selected poisonings, PyTorch model had specificity of 97%, accuracy of 83%, precision of 83%, recall of 83%, and a F1-score of 82%. Keras had specificity of 98%, accuracy of 83%, precision of 84%, recall of 83%, and a F1-score of 83%. The best performance was achieved in the diagnosis of single-agent poisoning in diagnosing poisoning by lithium, sulfonylureas, diphenhydramine, calcium channel blockers, then acetaminophen, in PyTorch (F1-score = 99%, 94%, 85%, 83%, and 82%, respectively) and Keras (F1-score = 99%, 94%, 86%, 82%, and 82%, respectively).

CONCLUSION

Deep neural networks can potentially help in distinguishing the causative agent of acute poisoning. This study used a small list of drugs, with polysubstance ingestions excluded.Reproducible source code and results can be obtained at https://github.com/ashiskb/npds-workspace.git.

Evolving kernel extreme learning machine for medical diagnosis via a disperse foraging sine cosine algorithm

Kernel extreme learning machine (KELM) has been widely used in the fields of classification and identification since it was proposed. As the parameters in the KELM model have a crucial impact on performance, they must be optimized before the model can be applied in practical areas. In this study, to improve optimization performance, a new parameter optimization strategy is proposed, based on a disperse foraging sine cosine algorithm (DFSCA), which is utilized to force some portions of search agents to explore other potential regions. Meanwhile, DFSCA is integrated into KELM to establish a new machine learning model named DFSCA-KELM. Firstly, using the CEC2017 benchmark suite, the exploration and exploitation capabilities of DFSCA were demonstrated. Secondly, evaluation of the model DFSCA-KELM on six medical datasets extracted from the UCI machine learning repository for medical diagnosis proved the effectiveness of the proposed model. At last, the model DFSCA-KELM was applied to solve two real medical cases, and the results indicate that DFSCA-KELM can also deal with practical medical problems effectively. Taken together, these results show that the proposed technique can be regarded as a promising tool for medical diagnosis.

Metabolic Changes in Mouse Plasma after Acute Diquat Poisoning by UPLC-MS/MS

Introduction:

Diquat is a fast-acting contact herbicide and plant dehydrating agent. The oral lethal dose 50 (LD50) of diquat in mice is about 125 mg/kg. The purpose of this study is to research the metabolomics in mouse plasma after acute diquat poisoning.

Method:

These mice were divided into two groups (the control group and acute diquat poisoning group). The control group was given normal saline by gavage. The acute diquat poisoning group was given 50 mg/kg diquat. UPLC-MS/MS was used to determinate the small molecule organic acid in mouse plasma.

Results:

Compare to the control group, the L-lysine, Adenine, L-Alanine, L-Valine, Lactic acid, Inosine, Adenosine, LTryptophan, L-Tyrosine, L-Arginine, L-Phenylalanine, L-Methionine, Citric acid, Fructose, L-Glutamine, Malic acid, LAspartic acid and Pyruvic acid increased in the acute diquat poisoning group (p<0.05); while the L-Histidine decreased (p<0.05).

Conclusion:

The results of metabolites increased or decreased, indicating that acute diquat poisoning induced amino acid metabolism and energy metabolism perturbations in mice.

Metabolic Changes in Rat Plasma After Epilepsy by UPLC-MS/MS

Introduction:

Epilepsy is one of the most common neurological diseases in clinical practice. The combined application of metabolomics technology plays a great advantage in the screening of biomarkers.

Methods:

In this study, Wistar rats were used as experimental subjects to model intractable epilepsy and to detect the metabolic changes of small molecules in plasma. UPLC-MS/MS was used to determine the small molecules in rat plasma. UPLC HSS C18 (2.1mm×100mm, 1.7 μm) column was used for separation, column temperature of 40°C. The initial mobile phase was acetonitrile -0.3% formic acid with gradient elution, the flow rate was 0.3 mL/min, total running time 4.0 min. Quantitative analysis was performed with multi-response monitoring (MRM).

Results:

Compared to the control group, the L-Alanine and L-Arginine decreased in the Epilepsy group (p<0.05); while Cytosine, Adenosine, L-Tyrosine, Citric acid, Fructose increased (p<0.05).

Conclusion:

In the screening of epilepsy biomarkers using metabolomics, various amino acids that lead to increased energy production and neurotransmitter imbalance play an important role in epileptic seizures.

Functional Metabolomics and Chemoproteomics Approaches Reveal Novel Metabolic Targets for Anticancer Therapy

Cancer cells exhibit different metabolic patterns compared to their normal counterparts. Although the reprogrammed metabolism has been indicated as strong biomarkers of cancer initiation and progression, increasing evidences suggest that metabolic alteration tuned by oncogenic drivers contributes to the occurrence and development of cancers rather than just being a hallmark of cancer. With this notion, targeting cancer metabolism holds promise as a novel anticancer strategy and is embracing its renaissance during the past two decades. Herein we have summarized the most recent developments in omics technology, including both metabolomics and proteomics, and how the combined use of these analytical tools significantly impacts this field by comprehensively and systematically recording the metabolic changes in cancer and hence reveals potential therapeutic targets that function by modulating the disrupted metabolic pathways.

Pharmacokinetics of Picroside I, II, III, IV in Rat Plasma by UPLCMS/ MS

Background:

Modern pharmacological studies show that rhizoma coptidis has protective effects on the liver, gallbladder, kidney, cerebral ischemia-reperfusion, local hypoxia injury, antiinflammatory, bone injury, nerve cells and myocardial cells. The effective components have been isolated from picroside I, II, III and IV.

Introduction:

A selective and sensitive ultra-performance liquid chromatography electrospray ionization tandem mass spectrometry (UPLC-ESI-MS/MS) method was developed for the simultaneous quantitative determination of picroside I, II, III and IV in rat plasma to aid the pharmacokinetics studies.

Method:

Sprague-Dawley (SD) rats were orally administered with 10 mg/kg, intravenously injected with 1 mg/kg for the mixture of picroside I, II, III and IV. The biological samples were collected at 0.083 3 h, 0.25 h, 1 h, 2 h, 4 h, 6 h, 8 h, 12 h, 24 h. A UPLC BEH C18 column (2.1 mm×50 mm, 1.7 μm) was used for chromatographic separation with the mobile phase consisting of acetonitrile and 0.1% formic acid by gradient elution. The flow rate was 0.4 mL/min. Multiple reaction monitoring (MRM) transitions were m/z 491.1→147.1 for picroside I, m/z 511.1→234.9 for picroside II, m/z 537.3→174.8 for picroside III and m/z 507.3→163.1 for picroside IV in negative ion mode.

Result:

The inter-day precision was less than 13%, the intra-day precision was less than 15%. The accuracy ranged from 89.4% to 111.1%. Recovery was higher than 79.1%, and the matrix effect ranged from 96.2% to 109.0%.

Conclusion:

The sensitive, rapid and selective UPLC-MS/MS method can be applied to the pharmacokinetic study of picroside I, II, III and IV in rats.