1. Universidad San Pablo-CEU

The Lipid Annotation Service (LAS) is a representational state transfer (REST) application programming interface (API) service designed to aid researchers performing lipid annotation. It assigns certainty levels (very unlikely, unlikely, likely, and very likely) to the putative annotations received as input and explains the rationale of such assignments. Its rules, obtained from the Centre for Metabolomics and Bioanalysis (CEMBIO) and from a literature review, enable LAS to extract evidence to support or refute the annotations automatically by checking the inter-rule relationships. LAS is the first metabolite annotation tool capable of explaining in natural language (English) the evidence that supports or refutes the annotations. This facilitates the understanding of the results by the user and, thus, increases the user's confidence in the results. Concerning its performance, in an evaluation of blood plasma samples whose compounds had previously been identified using well-established standards, LAS yielded an F-measure higher than 80%.

Retention time information is used for metabolite annotation in metabolomic experiments. But its usefulness is hindered by the availability of experimental retention time data in metabolomic databases, and by the lack of reproducibility between different chromatographic methods. Accurate prediction of retention time for a given chromatographic method would be a valuable support for metabolite annotation. We have trained state-of-the-art machine learning regressors using the 80, 038 experimental retention times from the METLIN Small Molecule Retention Tim (SMRT) dataset. The models included deep neural networks, deep kernel learning, several gradient boosting models, and a blending approach. 5, 666 molecular descriptors and 2, 214 fingerprints (MACCS166, Extended Connectivity, and Path Fingerprints fingerprints) were generated with the alvaDesc software. The models were trained using only the descriptors, only the fingerprints, and both types of features simultaneously. Bayesian hyperparameter search was used for parameter tuning. To avoid data-leakage when reporting the performance metrics, nested cross-validation was employed. The best results were obtained by a heavily regularized deep neural network trained with cosine annealing warm restarts and stochastic weight averaging, achieving a mean and median absolute errors of 39.2 ± 1.2 s and 17.2 ± 0.9 s , respectively. To the best of our knowledge, these are the most accurate predictions published up to date over the SMRT dataset. To project retention times between chromatographic methods, a novel Bayesian meta-learning approach that can learn from just a few molecules is proposed. By applying this projection between the deep neural network retention time predictions and a given chromatographic method, our approach can be integrated into a metabolite annotation workflow to obtain z-scores for the candidate annotations. To this end, it is enough that just as few as 10 molecules of a given experiment have been identified (probably by using pure metabolite standards). The use of z-scores permits considering the uncertainty in the projection when ranking candidates, and not only the accuracy. In this scenario, our results show that in 68% of the cases the correct molecule was among the top three candidates filtered by mass and ranked according to z-scores. This shows the usefulness of this information to support metabolite annotation. Python code is available on GitHub at https://github.com/constantino-garcia/cmmrt.

Retention time information is used for metabolite annotation in metabolomic experiments. But its usefulness is hindered by the availability of experimental retention time data in metabolomic databases, and by the lack of reproducibility between different chromatographic methods. Accurate prediction of retention time for a given chromatographic method would be a valuable support for metabolite annotation. We have trained state‑of‑the‑art machine learning regressors using the 80, 038 experimental retention times from the METLIN Small Molecule Retention Tim (SMRT) dataset. The models included deep neural networks, deep kernel learning, several gradient boosting models, and a blending approach. 5, 666 molecular descriptors and 2, 214 fingerprints (MACCS166, Extended Connectivity, and Path Fingerprints fingerprints) were generated with the alvaDesc software. The models were trained using only the descriptors, only the fingerprints, and both types of features simultaneously. Bayesian hyperparameter search was used for parameter tuning. To avoid data‑leakage when reporting the performance metrics, nested cross‑validation was employed. The best results were obtained by a heavily regularized deep neural network trained with cosine annealing warm restarts and stochastic weight averaging, achieving a mean and median absolute errors of 39.2 ± 1.2 s and 17.2 ± 0.9 s , respectively. To the best of our knowledge, these are the most accurate predictions published up to date over the SMRT dataset. To project retention times between chromatographic methods, a novel Bayesian meta‑learning approach that can learn from just a few molecules is proposed. By applying this projection between the deep neural network retention time predictions and a given chromatographic method, our approach can be integrated into a metabolite annotation workflow to obtain z‑scores for the candidate annotations. To this end, it is enough that just as few as 10 molecules of a given experiment have been identified (probably by using pure metabolite standards). The use of z‑scores permits considering the uncertainty in the projection when ranking candidates, and not only the accuracy. In this scenario, our results show that in 68% of the cases the correct molecule was among the top three candidates filtered by mass and ranked according to z‑scores. This shows the usefulness of this information to support metabolite annotation. Python code is available on GitHub at https://github.com/ constantino‑garcia/cmmrt.

The mechanisms driving SARS-CoV-2 susceptibility remain poorly understood, especially the factors determining why unvaccinated individuals remain uninfected despite high-risk exposures. To understand lipid and metabolite profiles related with COVID-19 susceptibility and disease progression. We collected samples from an exceptional group of unvaccinated healthcare workers heavily exposed to SARS-CoV-2 but not infected (‘non-susceptible’) and subjects who became infected during the follow-up (‘susceptible’), including non-hospitalized and hospitalized patients with different disease severity providing samples at early disease stages. Then, we analyzed their plasma metabolomic profiles using mass spectrometry coupled with liquid and gas chromatography. We show specific lipids profiles and metabolites that could explain SARS-CoV-2 susceptibility and COVID-19 severity. More importantly, non-susceptible individuals show a unique lipidomic pattern characterized by the upregulation of most lipids, especially ceramides and sphingomyelin, which could be interpreted as markers of low susceptibility to SARS-CoV-2 infection. This study strengthens the findings of other researchers about the importance of studying lipid profiles as relevant markers of SARS-CoV-2 pathogenesis.

Ceu Mass Mediator (CMM - http://ceumass.eps.uspceu.es/) is an on-line tool that has evolved from a simple interface to query di erent metabolomic databases (CMM1.0) to a tool that uni es the compounds from these databases and, using an expert system with knowledge about the experimental set-up and the compounds properties, lters and scores the query results (CMM 2.0). Since this last major revision, CMM has continued to grow, expanding the knowledge base of its expert system and including new services to support researchers in the metabolite annotation and identi cation process. The information from external databases has been refreshed and an in-house library with oxidized lipids not present in other sources has been added. This has increased the number of experimental metabolites up to 332,665 and the number of predicted metabolites up to 681,198. Furthermore, new taxonomy and ontology metadata have been included. CMM has expanded its functionalities with a service for the annotation of oxidized glycerophosphocholines, a service for spectral comparison from MS2 data and a spectral quality assessment service to determine the reliability of a spectrum for compound identi cation purposes. To facilitate the collaboration and integration of CMM with external tools and metabolomic platforms, a RESTful API has been created and it has already been integrated into the HMDB (Human Metabolome Database). This paper will present the novel functionalities incorporated to the version 3.0 of CMM.

Oxylipins are signaling lipids derived from the oxidation of polyunsaturated fatty acids (PUFAs). In lipidomic studies, human plasma may be subjected to various storage conditions and freeze-thaw cycles, which may impact the analysis of these compounds. In this study, we used liquid chromatography coupled with mass spectrometry (LC-MS) to examine the influence of up to five freeze-thaw cycles (FTCs) on free and total (mostly esterified) oxylipins in human plasma and the influence of temperature and storage duration (4 °C for up to 120 h and –20 °C and –80 °C for 1–98 days) in the presence or absence of butylated hydroxytoluene (BHT) on extracted oxylipins stored in LC-MS amber vials. In fresh plasma subjected to several FTCs, approximately 48 % of the detected free oxylipins were significantly altered by the third cycle, with increases in cytochrome P450 (CYP450) and lipoxygenase (LOX)-derived compounds and reductions in trihydroxylated oxylipins. In contrast, multiple FTCs did not significantly alter esterified oxylipins. At 4 °C, the extracted oxylipins did not change significantly for up to 120 h (5 days). Oxylipin levels remained stable for 98 days at –80 °C but decreased by 98 days at –20 °C. The antioxidant activity of butylated hydroxytoluene (BHT) did not influence oxylipin stability at 4 °C for 120 h or at –80 °C for 98 days, but it reduced oxylipin degradation at –20 °C at 98 days. Conversely, prostaglandin F2α (PGF2α) exhibited substantial increases at –20 °C and –80 °C, independent of BHT. This study demonstrates that (i) unlike free oxylipins, the esterified oxylipin pool remains stable following repeated FTCs, (ii) extracted oxylipins are stable at 4 °C for up to 120 h and at –80 °C for up to 98 days, but not at –20 °C for 98 days, and (iii) BHT may minimize oxylipin degradation of sample extracts stored at –20 °C. This study provides a framework for measuring oxylipins under various freeze-thaw and storage conditions.

Syrian hamsters undergo a reversible hyperphosphorylation of protein τ during hibernation, providing a unique natural model that may unveil the physiological mechanisms behind this critical process involved in the development of Alzheimer’s disease and other tauopathies. The hibernation cycle of these animals fluctuates between a pair of stages: 3–4 days of torpor bouts interspersed with periods of euthermia called arousals that last several hours. In this study, we investigated for the first time the metabolic changes in brain tissue during hibernation. A total of 337 metabolites showed statistically significant differences during hibernation. Based on these metabolites, several pathways were found to be significantly regulated and, therefore, play a key role in the regulation of hibernation processes. The increase in the levels of ceramides containing more than 20 C atoms was found in torpor animals, reflecting a higher activity of CerS2 during hibernation, linked to neurofibrillary tangle generation and structural changes in the Golgi apparatus. Our results open up the debate about the possible significance of some metabolites during hibernation, which may possibly be related to τ phosphorylation and dephosphorylation events. In general, this study may provide insights into novel neuroprotective agents because the alterations described throughout the hibernation process are reversible.

Recombinant human growth hormone (rhGH) therapy is the primary treatment for children born small for gestational age (SGA) who fail to show spontaneous catch-up growth by two or four years. While its effects on white adipose tissue are well-documented, this pilot study aimed to investigate its impact on the lipidome and the thermogenic and endocrine activities of brown adipose tissue (BAT) in SGA children following rhGH treatment. The study involved 11 SGA children divided into two groups: a) SGA children who were not treated with rhGH (n=4) and b) SGA children who received rhGH treatment with Saizen® (n=7). This second group of seven SGA children was followed for 12 months after initiating rhGH treatment. Interventions included 12-hour fasting blood extraction and infrared thermography at baseline and 3 and 12 months post-treatment. Five appropriate-for-gestational-age (AGA) children served as controls. Exclusion criteria included endocrinological, genetic, or chronic diseases. Untargeted lipidomics analysis was performed using liquid chromatography-mass spectrometry (LC-MS), and serum biomarker levels were measured using ELISA assays. Serum lipidomic analysis revealed that free fatty acids (FFAs) increased to levels close to those of the AGA group after three months of rhGH administration, including polyunsaturated fatty acids, correlating with reduced leptin levels. Elevated levels of 1a,1b-dihomo-PGJ2 and adrenic acid suggested potential aging markers. rhGH treatment also significantly reduced meteorin-like (METRNL) and monocyte chemoattractant protein-1 (MCP1) serum levels to control levels. rhGH influences the serum lipidome, promoting changes in maturation and metabolism. Further research is required to clarify the direct effects of rhGH on specific lipid species and batokines, potentially addressing metabolic disturbances linked to obesity and aging.

Background: The pathological mechanism of the gastrointestinal forms of food aller-gies is less understood in comparison to other clinical phenotypes, such as asthmaand anaphylaxis Importantly, high-IgE levels are a poor prognostic factor in gastroin-testinal allergies.Methods: This study investigated how high-IgE levels influence the development ofintestinal inflammation and the metabolome in allergic enteritis (AE), using IgE knock-in (IgEki) mice expressing high levels of IgE. In addition, correlation of the altered me-tabolome with gut microbiome was analysed.Results: Ovalbumin-sensitized and egg-white diet-fed (OVA/EW) BALB/c WT micedeveloped moderate AE, whereas OVA/EW IgEki mice induced more aggravated in-testinal inflammation with enhanced eosinophil accumulation. Untargeted metabo-lomics detected the increased levels of N-tau-methylhistamine and 2,3-butanediol,and reduced levels of butyric acid in faeces and/or sera of OVA/EW IgEki mice, whichwas accompanied with reduced Clostridium and increased Lactobacillus at the genus level. Non-sensitized and egg-white diet-fed (NC/EW) WT mice did not exhibit anysigns of AE, whereas NC/EW IgEki mice developed marginal degrees of AE. Comparedto NC/EW WT mice, enhanced levels of lysophospholipids, sphinganine and sphin-gosine were detected in serum and faecal samples of NC/EW IgEki mice. In addi-tion, several associations of altered metabolome with gut microbiome—for exampleAkkermansia with lysophosphatidylserine—were detected.Conclusions: Our results suggest that high-IgE levels alter intestinal and systemic levelsof endogenous and microbiota-associated metabolites in experimental AE. This studycontributes to deepening the knowledge of molecular mechanisms for the developmentof AE and provides clues to advance diagnostic and therapeutic strategies of allergicdiseases