Comprehensive Lipidomic Automation Workflow using Large Language Models

Lipidomics generates large data that makes manual annotation and interpretation challenging. Lipid chemical and structural diversity with structural isomers further complicates annotation. Although, several commercial and open-source software for targeted lipid identification exists, it lacks automated method generation workflows and integration with statistical and bioinformatics tools. We have developed the Comprehensive Lipidomic Automated Workflow (CLAW) platform with integrated workflow for parsing, detailed statistical analysis and lipid annotations based on custom multiple reaction monitoring (MRM) precursor and product ion pair transitions. CLAW contains several modules including identification of carbon-carbon double bond position(s) in unsaturated lipids when combined with ozone electrospray ionization (OzESI)-MRM methodology. To demonstrate the utility of the automated workflow in CLAW, large-scale lipidomics data was collected with traditional and OzESI-MRM profiling on biological and non-biological samples. Specifically, a total of 1497 transitions organized into 10 MRM-based mass spectrometry methods were used to profile lipid droplets isolated from different brain regions of 18-24 month-old Alzheimer's disease mice and age-matched wild-type controls. Additionally, triacyclglycerols (TGs) profiles with carbon-carbon double bond specificity were generated from canola oil samples using OzESI-MRM profiling. We also developed an integrated language user interface with large language models using artificially intelligent (AI) agents that permits users to interact with the CLAW platform using a chatbot terminal to perform statistical and bioinformatic analyses. We envision CLAW pipeline to be used in high-throughput lipid structural identification tasks aiding users to generate automated lipidomics workflows ranging from data acquisition to AI agent-based bioinformatic analysis.

Lipid Isobar and Isomer Imaging Using Nanospray Desorption Electrospray Ionization Combined with Triple Quadrupole Mass Spectrometry

Mass spectrometry imaging (MSI) is widely used for examining the spatial distributions of molecules in biological samples. Conventional MSI approaches, in which molecules extracted from the sample are distinguished based on their mass-to-charge ratio, cannot distinguish between isomeric species and some closely spaced isobars. To facilitate isobar separation, MSI is typically performed using high-resolution mass spectrometers. Nevertheless, the complexity of the mixture of biomolecules observed in each pixel of the image presents a challenge, even for modern mass spectrometers with the highest resolving power. Herein, we implement nanospray desorption electrospray ionization (nano-DESI) MSI on a triple quadrupole (QqQ) mass spectrometer for the spatial mapping of isobaric and isomeric species in biological tissues. We use multiple reaction monitoring acquisition mode (MRM) with unit mass resolution to demonstrate the performance of this new platform by imaging lipids in mouse brain and rat kidney tissues. We demonstrate that imaging in MRM mode may be used to distinguish between isobaric phospholipids requiring a mass resolving power of 3,800,000. Additionally, we have been able to image eicosanoid isomers, a largely unexplored class of signaling molecules present in tissues at low concentrations, in rat kidney tissue. This new capability substantially enhances the specificity and selectivity of MSI, enabling spatial localization of species that remain unresolved in conventional MSI experiments.

Digital Quadrupole Isolation and Electron Capture Dissociation on an Extended Mass Range Q-TOF Provides Sequence and Structure Information on Proteins and Protein Complexes

Electron capture dissociation (ECD) is now a well-established method for sequencing peptides and performing top-down analysis on proteins of less than 30 kDa, and there is growing interest in using this approach for studies of larger proteins and protein complexes. Although much progress on ECD has been made over the past few decades, establishing methods for obtaining informative spectra still poses a significant challenge. Here we describe how digital quadrupole (DigiQ) ion isolation can be used for the mass selection of single charge states of proteins and protein complexes prior to undergoing ECD and/or charge reduction. First, we demonstrate that the DigiQ can isolate single charge states of monomeric proteins such as ubiquitin (8.6 kDa) and charge states of large protein complexes such as pyruvate kinase (234 kDa) using a hybrid quadrupole-TOF-MS (Agilent extended m/z range 6545XT). Next, we demonstrate that fragment ions resulting from ECD can be utilized to provide information about the sequence and structure of the cytochrome c/heme complex and the ubiquitin monomer. Lastly, an especially interesting result for DigiQ isolation and electron capture (EC) was noted; namely, the 16+ charge state of the streptavidin/biotin complex reveals different electron capture patterns for the biotinylated proteoforms of streptavidin. This result is consistent with previous reports that apo streptavidin exists in multiple conformations and that biotin binding shifts the conformational dynamics of the complex (Quintyn, R. Chem. Biol. 2015, 22 (55), 583-592).

3DMolMS: Prediction of Tandem Mass Spectra from Three Dimensional Molecular Conformations

MOTIVATION

Tandem mass spectrometry is an essential technology for characterizing chemical compounds at high sensitivity and throughput, and is commonly adopted in many fields. However, computational methods for automated compound identification from their MS/MS spectra are still limited, especially for novel compounds that have not been previously characterized. In recent years, in silico methods were proposed to predict the MS/MS spectra of compounds, which can then be used to expand the reference spectral libraries for compound identification. However, these methods did not consider the compounds' three-dimensional (3D) conformations, and thus neglected critical structural information.

RESULTS

We present the 3D Molecular Network for Mass Spectra Prediction (3DMolMS), a deep neural network model to predict the MS/MS spectra of compounds from their 3D conformations. We evaluated the model on the experimental spectra collected in several spectral libraries. The results showed that 3DMolMS predicted the spectra with the average cosine similarity of 0.691 and 0.478 with the experimental MS/MS spectra acquired in positive and negative ion modes, respectively. Furthermore, 3DMolMS model can be generalized to the prediction of MS/MS spectra acquired by different labs on different instruments through minor fine-tuning on a small set of spectra. Finally, we demonstrate that the molecular representation learned by 3DMolMS from MS/MS spectra prediction can be adapted to enhance the prediction of chemical properties such as the elution time (ET) in the liquid chromatography and the collisional cross section (CCS) measured by ion mobility spectrometry (IMS), both of which are often used to improve compound identification.

AVAILABILITY

The codes of 3DMolMS are available at https://github.com/JosieHong/3DMolMS and the web service is at https://spectrumprediction.gnps2.org.

Ion Manipulation in Open Air Using 3D-Printed Electrodes

Ambient ionization techniques provide a way to sample materials via creation of ions in the air. However, transferring and focusing of these ions is typically done in the reduced pressure environment of the mass spectrometer. Spray-based ambient ionization sources require relatively large distances between the source and mass spectrometer inlet for effective desolvation, resulting in a small fraction of the ions being collected. To increase the efficiency of ion transfer from atmosphere to vacuum, 3D-printed focusing devices made of conductive carbon nanotube doped polymers have been designed and evaluated for ion focusing in air. Three main classes of electrodes are considered: (i) conic section electrodes (conical, ellipsoidal, and cylindrical), (ii) simple conductive and non-conductive apertures, and (iii) electrodes with complex geometries (straight, chicane, and curved). Simulations of ion trajectories performed using the statistical diffusion simulation (SDS) model in SIMION showed a measure of agreement with experiment. Cross-sectional images of ion beams were captured using an ion detecting charge-coupled device (IonCCD). After optimization, the best arrangements of electrodes were coupled to an Agilent Ultivo triple quadrupole to record mass spectra. Observations suggest that electrode geometry strongly influences ion trajectories in air. Non-conductive electrodes also assisted in focusing, due to charge buildup from ion deposition. We also observed minimal spreading of the ion packet after exiting the focusing electrodes indicating that atmospheric collisions do not reduce collimation of the beam. The study suggests that high pressures need not be viewed as a hindrance to ion transport, but as a potentially useful force.

Benzyl benzoate glycosides from Oligoneuron rigidum

Two new benzyl benzoate glycosides were isolated from the leaves and stems of the native North American prairie plant Oligoneuron rigidum (stiff goldenrod). The glycosides were isolated as a mixture of benzoate and acetate esters, which were subjected to mild base hydrolysis to facilitate full structural characterization using LCMSMS and 1D and 2D NMR data.

[Comparative study of the efficiency of bacterial filters in long-term mechanical ventilation]

Two commercially available bacterial filters to be used as part of the mechanical ventilation unit during anaesthesia were tested for hygienic criteria. Manufacturers claim that bacterial breathing filters have a filtration capacity of about 99.995%, so that there would be no need for thermal disinfection of tubing and ventilation circuits after each use. One filter is designed for a single use only, the other can be used up to 24 times after sterilisation. Both filters consist of hydrophobic glass fibres.

METHODS

During simulated mechanical ventilation for 24 h, an alcoholic suspension of Bacillus subtilis was atomized in front of the filters tested. A gelatin membrane filter was integrated in the ventilation circuit and captured the filtered gas behind the test filter.

RESULTS

During simulated mechanical ventilation for 24 h, the filtration capacity of both the disposable and reusable filters (Table 2) did not confirm the manufacturers' short-term technical findings over 8 s (DIN-EN 143).

CONCLUSIONS

The use of bacterial filters during mechanical ventilation reduces the probability of bacterial contamination, but does not make sterilisation of the tubes and ventilation circuit unnecessary.