Experimental Collision Cross Sections Research Articles

Ion Mobility-Mass Spectrometry Reveals the Structures and Stabilities of Biotherapeutic Antibody Aggregates.

Stability is a key critical quality attribute monitored throughout the development of monoclonal antibody (mAb) therapeutics. Minor changes in their higher order structure (HOS) caused by stress or environment may alter mAb aggregation, immunogenicity, and efficacy. In addition, the structures of the resulting mAb aggregates are largely unknown, as are their dependencies on conditions under which they are created. In this report, we investigate the HOS of mAb monomers and dimers under a variety of forced degradation conditions with ion mobility-mass spectrometry (IM-MS) and collision-induced unfolding (CIU) technologies. We evaluate two model IgG1 antibodies that differ significantly only in their complementarity-determinant regions: IgG1α and IgG1β. Our data covering both heat- and pH-based forced degradation conditions, aquired on two different IM-MS platforms, show that these mAbs undergo global HOS changes at both monomer and dimer levels upon degradation, but shifts in collision cross section (CCS) differ under pH or heat degradation conditions. In addition, the level of CCS change detected is different between IgG1α and IgG1β, suggesting that differences in the CDR drive differential responses to degradation that influence the antibody HOS. Dramatically different CIU fingerprints are obtained for IgG1α and IgG1β monomers and dimers for both degradation conditions. Finally, we constructed a series of computational models of mAb dimers for comparison with experimental CCS values and found evidence for a compact, overlapped dimer structure under native and heat degradation conditions, possibly adopting an inverted or nonoverlapped quaternary structure when produced through pH degredation. We conclude by discussing the potential impact of our findings on ongoing biotherapeutic discovery and development efforts.

Molecular networking and collision cross section prediction for structural isomer and unknown compound identification in plant metabolomics: a case study applied to Zhanthoxylum heitzii extracts

Mass spectrometry-based plant metabolomics allow large-scale analysis of a wide range of compounds and the discovery of potential new active metabolites with minimal sample preparation. Despite recent tools for molecular networking, many metabolites remain unknown. Our objective is to show the complementarity of collision cross section (CCS) measurements and calculations for metabolite annotation in a real case study. Thus, a systematic and high-throughput investigation of root, bark, branch, and leaf of the Gabonese plant Zhanthoxylum heitzii was performed through ultra-high performance liquid chromatography high-resolution tandem mass spectrometry (UHPLC-QTOF/MS). A feature-based molecular network (FBMN) was employed to study the distribution of metabolites in the organs of the plants and discover potential new components. In total, 143 metabolites belonging to the family of alkaloids, lignans, polyphenols, fatty acids, and amino acids were detected and a semi-quantitative analysis in the different organs was performed. A large proportion of medical plant phytochemicals is often characterized by isomerism and, in the absence of reference compounds, an additional dimension of gas phase separation can result in improvements to both quantitation and compound annotation. The inclusion of ion mobility in the ultra-high performance liquid chromatography mass spectrometry workflow (UHPLC-IMS-MS) has been used to collect experimental CCS values in nitrogen and helium (CCSN2 and CCSHe) of Zhanthoxylum heitzii features. Due to a lack of reference data, the investigation of predicted collision cross section has enabled comparison with the experimental values, helping in dereplication and isomer identification. Moreover, in combination with mass spectra interpretation, the comparison of experimental and theoretical CCS values allowed annotation of unknown features. The study represents a practical example of the potential of modern mass spectrometry strategies in the identification of medicinal plant phytochemical components.

Uncovering the behaviour of ions in the gas-phase to predict the ion mobility separation of isomeric steroid compounds

Bile acids are steroid compounds involved in biological mechanisms of neurodegenerative diseases making them potential biomarkers for diagnosis or treatment. These compounds exist as structural and conformational isomers, which hinder distinguishing them in physiological processes. We aimed to develop tandem mass spectrometry-ion mobility spectrometry (MS/MS-IMS) methodologies to explore and understand the behaviour of isomeric steroids in the gas-phase and rapidly separate them. Unlike previously published ion mobility data, various isomers were investigated in mixtures to better mimic complex (pre-) clinical samples. The experimental collision cross sections (CCS)s were compared to the theoretical CCS values for an in-depth analysis of isomeric ions' behaviour in the gas-phase. Based on density-functional theory, we identified the impact of adduct positioning on the 3D conformation of enantiomers, diastereomers and structural isomers. The curling of the large side chains hedged the small differences among the isomers and lowered the CCS values. On the other hand, fragmenting off the identical side branches as well as imposing the bending of the steroid ring resulted in ion mobility differentiation. Careful data evaluation revealed the tendency of isomers to form homo-cluster in the mixture solutions and assist the separation. Our fundamental and experimental findings enable the ion mobility separation of isomeric steroids to be predicted. The introduced rapid and optimal MS/MS-IMS analytical methodology can be applied to distinguish isomeric bile acids both in a solution and potentially in patients’ tissue samples, and consequently, reveal their molecular pathways.

Geometrical Structures of Gas-Phase Cerium Oxide Cluster Cations after Reaction with Nitric Oxide Studied by Ion Mobility Mass Spectrometry.

Cerium oxide cluster cations were reacted with nitric oxide molecules and then measured by ion mobility mass spectrometry (IMMS). CenO2n+1N+ species appeared as products of the reaction CenO2n+ + NO → CenO2n+1N+, and their collision cross sections (CCSs) with helium were obtained by IMMS. The experimental CCSs of CenO2n+1N+ were 2-6 Å2 larger than those of CenO2n+ for n = 4-10. Geometrical structures of Ce4O9N+ and Ce5O11N+ were assigned by comparing experimental CCSs with theoretically calculated CCSs of candidate structures. The suggested structures showed that the adsorbed NO molecule is oxidized by the CenO2n+ cluster into a nitrite (NO2-) or nitrate (NO3-). The CenO2n+1N+ species are regarded as intermediates of the NO oxidation reaction CenO2n+ + NO → CenO2n-1+ + NO2, and therefore, the present results are helpful for understanding redox reactions involving gas-phase CenO2n+ cluster ions.

Influence Exerted by the Solvent Effect on the Mobility Peak of 1,8-Naphthalic Anhydride in Ion Mobility Spectrometry.

The collision cross-section (CCS) values of ions determined by ion mobility-mass spectrometry (IM-MS) can be used to deduce the shape and size of the ions. For each compound, as well as its isomer or tautomer, a unique arrival time peak was obtained in extracted ion mobility (EIM) spectra, which corresponded to a specific CCS value. However, the generation of solvated ions by electrospray ionization (ESI) increases the number of mobility peaks, which makes the EIM spectra difficult to interpret. In this study, solvent clusters formed by acetonitrile and methanol around 1,8-naphthalic anhydride (1,8-NA) cations ([C12H6O3 + H]+1,8-NA) were investigated using trapped ion mobility spectrometry-time-of-flight mass spectrometry (TIMS-TOF MS). The effects of infusion flow rate, nebulizer gas pressure, drying gas rate, and drying gas temperature on the formation of solvent clusters from acetonitrile and methanolic solution were systematically studied. The formation of solvent clusters was observed with infusion flow rates increased, which was manifested by the larger experimental CCS values of [C12H6O3 + H]+1,8-NA. Acetonitrile tended to form solvent clusters around ions more readily than methanol. These solvent clusters were stable enough to be detected by TIMS, but they cannot survive under ion activation conditions of mass spectrometry (MS). Increasing the nebulizer gas pressure seems to be a better way to eliminate the formation of solvent clusters in TIMS-TOF MS and give a "cleaner" EIM spectra. The current research demonstrates that more attention should be paid to the solvent effect on CCS values and their interpretation.

Geometric Analysis of Shapes in Ion Mobility-Mass Spectrometry.

Experimental ion mobility-mass spectrometry (IM-MS) results are often correlated to three-dimensional structures based on theoretical chemistry calculations. The bottleneck of this approach is the need for accurate values, both experimentally and theoretically predicted. Here, we continue the development of the trend-based analyses to extract structural information from experimental IM-MS data sets. The experimental collision cross-sections (CCSs) of synthetic systems such as homopolymers and small ionic clusters are investigated in terms of CCS trends as a function of the number of repetitive units (e.g., degree of polymerization (DP) for homopolymers) and for each detected charge state. Then, we computed the projected areas of expanding but perfectly defined geometric objects using an in-house software called MoShade. The shapes were modeled using computer-aided design software where we considered only geometric factors: no atoms, mass, chemical potentials, or interactions were taken into consideration to make the method orthogonal to classical methods for 3D shape assessments using time-consuming computational chemistry. Our modeled shape evolutions favorably compared to experimentally obtained CCS trends, meaning that the apparent volume or envelope of homogeneously distributed mass effectively modeled the ion-drift gas interactions as sampled by IM-MS. The CCSs of convex shapes could be directly related to their surface area. More importantly, this relationship seems to hold even for moderately concave shapes, such as those obtained by geometry-optimized structures of ions from conventional computational chemistry methods. Theoretical sets of expanding beads-on-a-string shapes allowed extracting accurate bead and string dimensions for two homopolymers, without modeling any chemical interactions.

Multi-CRAFTI: Relative Collision Cross Sections from Fourier Transform Ion Cyclotron Resonance Mass Spectrometric Line Width Measurements.

Determination of collision cross sections (CCS) using the cross-sectional areas by the Fourier transform ion cyclotron resonance (CRAFTI) technique is limited by the requirement that accurate pressures in the trapping cell of the mass spectrometer must be known. Experiments must also be performed in the energetic hard-sphere regime such that ions decohere after single collisions with neutrals; this limits application to ions that are not much more massive than the neutrals. To mitigate these problems, we have resonantly excited two (or more) ions of different m/z to the same center-of-mass kinetic energy in a single experiment, subjecting them to identical neutral pressures. We term this approach "multi-CRAFTI". This facilitates measurement of relative CCS without requiring knowledge of the pressure and enables determination of absolute CCS using internal standards. Experiments with tetraalkylammonium ions yield CCS in reasonable agreement with the one-ion-at-a-time CRAFTI approach and with ion mobility spectrometry (IMS) when differences in collision energetics are taken into account (multi-CRAFTI generally yields smaller CCS than does IMS due to the higher collision energies employed in multi-CRAFTI). Comparison of multi-CRAFTI and IMS results with CCS calculated from structures computed at the M06-2X/6-31+G* level of theory using projection approximation or trajectory method values, respectively, indicates that the computed structures have CCS increasingly smaller than the experimental CCS as m/z increases, implying the computational model overestimates interactions between the alkyl arms. For ions that undergo similar collisional decoherence processes, relative CCS reach constant values at lower collision energies than do absolute CCS values, suggesting a means of increasing the accessible upper m/z limit by employing multi-CRAFTI.

Investigation into Small Molecule Isomeric Glucuronide Metabolite Differentiation Using In Silico and Experimental Collision Cross-Section Values.

Identifying isomeric metabolites remains a challenging and time-consuming process with both sensitivity and unambiguous structural assignment typically only achieved through the combined use of LC-MS and NMR. Ion mobility mass spectrometry (IMMS) has the potential to produce timely and accurate data using a single technique to identify drug metabolites, including isomers, without the requirement for in-depth interpretation (cf. MS/MS data) using an automated computational pipeline by comparison of experimental collision cross-section (CCS) values with predicted CCS values. An ion mobility enabled Q-Tof mass spectrometer was used to determine the CCS values of 28 (14 isomeric pairs of) small molecule glucuronide metabolites, which were then compared to two different in silico models; a quantum mechanics (QM) and a machine learning (ML) approach to test these approaches. The difference between CCS values within isomer pairs was also assessed to evaluate if the difference was large enough for unambiguous structural identification through in silico prediction. A good correlation was found between both the QM- and ML-based models and experimentally determined CCS values. The predicted CCS values were found to be similar between ML and QM in silico methods, with the QM model more accurately describing the difference in CCS values between isomer pairs. Of the 14 isomeric pairs, only one (naringenin glucuronides) gave a sufficient difference in CCS values for the QM model to distinguish between the isomers with some level of confidence, with the ML model unable to confidently distinguish the studied isomer pairs. An evaluation of analyte structures was also undertaken to explore any trends or anomalies within the data set.

Conformer Separation of Dibenzo-Crown-Ether Complexes with Na+ and K+ Ions Studied by Cryogenic Ion Mobility-Mass Spectrometry.

We performed cryogenic ion mobility-mass spectrometry (IM-MS) to study conformations of dibenzo-crown-ether complexes with Na+ and K+ ions at 86 K in the gas phase. Four dibenzo-crown-ethers (dibenzo-18-crown-6, dibenzo-21-crown-7, dibenzo-24-crown-8, and dibenzo-30-crown-10) with different cavity ring sizes were investigated. For dibenzo-18-crown-6 complexes with Na+ and K+, only one type of conformer was assigned by comparing the experimental collision cross sections with those predicted theoretically for candidate structures. In this conformer, the distance between two benzene rings in the complexes was long due to the open form of the dibenzo-18-crown-6. This open conformer was consistent with the previous laser spectroscopic studies of the cold complex ions in the gas phase. For dibenzo-21-crown-7 and dibenzo-24-crown-8 complexes with Na+ and K+, two types of conformers were clearly separated by IM-MS. These two conformer types were assigned to "open" and "closed" forms in which benzene-benzene distances were long and short, respectively. Observed relative abundances of the open and closed conformers qualitatively agreed with the Boltzmann distribution using Gibbs energies of the conformers calculated by quantum chemical calculations. For the Na+(dibenzo-30-crown-10) complex, open and closed conformers were also observed in IM-MS. On the other hand, only the closed conformer was observed for the K+(dibenzo-30-crown-10) complex. This closed conformer was similar to the "wraparound" structure, which was proposed in the previous studies in the solution. In conclusion, the closed conformers were formed by the deformation of flexible crown ethers with large cavity ring sizes. In addition, the diameter of the K+ ion was suitable to form the closed conformer by deformation of the molecular structure of dibenzo-30-crown-10.

Effective collisional cross-section of small ions in the gas phase: Application to ion mobility spectrometry.

The observed drift times of monoatomic ions, including alkali metal ions and halide anions, are not fully consistent with their size. When the effect of mass is included through the Mason-Schamp equation, the deviation gets worse so that the trend of the experimental collisional cross-sections becomes completely opposite to what is expected. This is attributed to the stronger local electric field around smaller ions. The strong electric field in the vicinity of a small ion leads to strong ion-neutral interactions and creates a drag force against ion motion. The smaller the ions, the stronger the interaction, because of the higher charge density. In view of this, a modified equation is introduced to describe the relationship between the observed drift times or ion mobilities and the cross-sections of small ions. Here, for small ions with high charge density, the experimental collision cross-section is expressed as the effective collision cross-section, Ωeff = σi (1 + α/ri 3 ), that takes into account both intrinsic ion size, σi , and the ion-molecule interactions through a correction term of α/ri 3 , which is proportional to the charge density. A linear fit of the drift times of alkali metal ions and halide anions to the proposed equation showed relative deviations of <8.2%. The model successfully predicted the drift time of other small diatomic ions with reasonable error. The proposed model can be used as a simple and efficient relationship in predicting the effective cross-section of small ions.

Localization and Composition of Fructans in Stem and Rhizome of Agave tequilana Weber var. azul.

Methodology combining mass spectrometry imaging (MSI) with ion mobility separation (IMS) has emerged as a biological imaging technique due to its versatility, sensitivity and label-free approach. This technique has been shown to separate isomeric compounds such as lipids, amino acids, carboxylic acids and carbohydrates. This report describes mass spectrometry imaging in combination with traveling-wave ion mobility separation and matrix-assisted laser desorption/ionization (MALDI). Positive ionization mode was used to locate fructans on tissue printed sections of Agave rhizome and stem tissue and distinguished fructan isoforms. Here we show the location of fructans ranging from DP3 to DP17 to be differentially abundant across the stem tissue and for the first time, experimental collision cross sections of endogenous fructan structures have been collected, revealing at least two isoforms for fructans of DP4, DP5, DP6, DP7, DP8, DP10, and DP11. This demonstrates that complex fructans such as agavins can be located and their isoforms resolved using a combination of MALDI, IMS, and MSI, without the need for extraction or derivatization. Use of this methodology uncovered patterns of fructan localization consistent with functional differences where higher DP fructans are found toward the central section of the stem supporting a role in long term carbohydrate storage whereas lower DP fructans are concentrated in the highly vascularized central core of rhizomes supporting a role in mobilization of carbohydrates from the mother plant to developing offsets. Tissue specific patterns of expression of genes encoding enzymes involved in fructan metabolism are consistent with fructan structures and localization.

Intramolecular Dispersion Attraction in Tetraalkylammonium Cations Revealed by Cryogenic Ion Mobility Mass Spectrometry.

We performed cryogenic ion mobility mass spectrometry and quantum chemical calculations of tetraalkylammonium (TAA) cations, (CnH2n+1)4N+ (n = 4-8), to study geometrical structures of TAA cations. We measured collision cross sections (CCSs) of TAA cations with He buffer gas atoms at 86 K. In addition, CCSs were calculated for optimized structures of TAA ions to compare with experimental CCSs. For n = 4 and 5, calculated CCSs of nearly planar conformers with all-trans four alkyl chains agree well with experimental CCSs. On the other hand, for n = 8, calculated CCSs of a conformer with two gauche alkyl chains reproduce experimental CCSs. The structural transition from all-trans to gauche conformers occurs around n = 6-8. The dispersion attraction between alkyl chains is a major interaction to stabilize the gauche conformer of n = 8.

First-Principles Collision Cross Section Measurements of Large Proteins and Protein Complexes.

Rotationally averaged collision cross section (CCS) values for a series of proteins and protein complexes ranging in size from 8.6 to 810 kDa are reported. The CCSs were obtained using a native electrospray ionization drift tube ion mobility-Orbitrap mass spectrometer specifically designed to enhance sensitivity while having high-resolution ion mobility and mass capabilities. Periodic focusing (PF)-drift tube (DT)-ion mobility (IM) provides first-principles determination of the CCS of large biomolecules that can then be used as CCS calibrants. The experimental, first-principles CCS values are compared to previously reported experimentally determined and computationally calculated CCS using projected superposition approximation (PSA), the Ion Mobility Projection Approximation Calculation Tool (IMPACT), and Collidoscope. Experimental CCS values are generally in agreement with previously reported CCSs, with values falling within ∼5.5%. In addition, an ion mobility resolution (CCS centroid divided by CCS fwhm) of ∼60 is obtained for pyruvate kinase (MW ∼ 233 kDa); however, ion mobility resolution for bovine serum albumin (MW ∼ 68 kDa) is less than ∼20, which arises from sample impurities and underscores the importance of sample quality. The high resolution afforded by the ion mobility-Orbitrap mass analyzer provides new opportunities to understand the intricate details of protein complexes such as the impact of post-translational modifications (PTMs), stoichiometry, and conformational changes induced by ligand binding.

Sodium Coordination and Protonation of Poly(ethoxy phosphate) Chains in the Gas Phase Probed by Ion Mobility-Mass Spectrometry.

The two-dimensional shape information yielded by ion mobility-mass spectrometry (IM-MS), usually reported as collision cross section (CCS), is often correlated to the underlying three-dimensional structures of the ions through computational chemistry. Here, we used theoretical approaches based on molecular mechanics (MM), molecular dynamics (MD), and density functional theory (DFT) to elucidate the structures of sodiated poly(ethoxy phosphate) polymer ions at different degrees of polymerization (DP) for three different charge states (1+, 2+, and 3+) by comparing computational results to experimentally obtained CCS values. From the calculated structures, we extract several key interaction distances which merge in clusters for all screened charge states and DPs, independent of the three-dimensional structures and the polymer ion structural rearrangements. These distances were also used to extract the minimum coordination numbers in poly(ethoxy phosphate) and to describe the preferred coordination geometries. When sodiated and protonated polymer ions are compared, the experimental CCS evolutions differ at small DP values and merge at higher DPs. We investigated in more depth this difference for two selected species, namely, [PEtP5 + 2Na+]2+ and [PEtP5 + 2H+]2+. For the protonated ions, we explored the different protonation sites to extract three-dimensional structure candidates and rationalize the CCS behaviors.

Conformational Characterization of Polyelectrolyte Oligomers and Their Noncovalent Complexes Using Ion Mobility-Mass Spectrometry.

Poly-l-lysine (PLL), polystyrenesulfonate (PSS), and a mixture of these polyelectrolytes were investigated by electrospray ionization ion mobility mass spectrometry. The IM step confirmed the formation of noncovalent (i.e., supramolecular) complexes between these polyelectrolytes, which were detected in various charge states and stoichiometries in the presence of their constituents. Experimental and theoretical collision cross sections (CCSs) were derived for both PLL and PSS oligomers as well as their noncovalent assemblies. PSS chains showed higher compactness with increasing size as compared to PLL chains, indicating that the intrinsic conformations of the polyelectrolytes depend on the nature of the functional groups on their side chains. The CCS data for the noncovalent complexes further revealed that assemblies with higher PLL content have higher CCS values than other compositions of similar mass and that PLL-PSS complex formation is accompanied by significant size contraction.

Structures of Magnesium Oxide Cluster Cations Studied Using Ion Mobility Mass Spectrometry.

Structural assignments of gas-phase magnesium oxide cluster cations, MgnOn+ (n ≤ 24), have been achieved from a comparison of experimental collision cross sections (CCSs) measured using ion mobility mass spectrometry and theoretical CCSs calculated for equilibrium structures optimized by quantum chemical calculations. Various structures based on rock-salt and hexagonal-tube structures were assigned for n = 5-13 and 15. On the other hand, only rock-salt type structures were assigned for n = 4, 14, 16-21, and 24. The CCS values and the total energies were close for the hexagonal-tube and rock-salt structures of a given size in the small size range (n ≤ 15 except for 4 and 14). The hexagonal-tube structures of the cluster ions with n ≥ 16 were less stable and had larger CCSs than the rock-salt structures. These results indicate that the structures of the MgnOn+ clusters were changed from mixtures of rock-salt and hexagonal-tube structures to pure rock-salt structures with the growth of the cluster size. All atoms in the hexagonal-tube structures are located on the surface of the clusters, even if the cluster size increases. In contrast, the assigned rock-salt structures with n = 13, 14, and n ≥ 17 had atoms inside the clusters, which means that the average coordination numbers are substantially higher for the rock-salt structures than for the hexagonal tube structures. The structural change from the mixed structures to pure rock-salt with an increase in size n can be attributed to this difference in the coordination number.

A Mechanistic Study of Protonated Aniline to Protonated Phenol Substitution Considering Tautomerization by Ion Mobility Mass Spectrometry and Tandem Mass Spectrometry.

We report the use of ion mobility mass spectrometry (IMMS) and energy-resolved collisional activation to investigate gas-phase reactions of protonated aniline and protonated phenol. Protonated aniline prototropic tautomerization and nucleophilic substitution (SN1) to produce phenol with traces of water in the IMMS cell are reported. Tautomerization of protonated phenol and its ability to form protonated aniline in presence of ammonia in the gas phase are also observed. These results are supported by energy landscapes obtained from computational chemistry. These structure modifications in the IMMS cell affected the measured collision cross section (CCS). A thorough understanding of the gas-phase reactions occurring in IMMS appears mandatory before using the experimental CCS as a robust descriptor which is stated by the recent literature.

Comparing Empty and Filled Fullerene Cages with High-Resolution Trapped Ion Mobility Spectrometry.

We have used trapped ion mobility spectrometry (TIMS) to obtain highly accurate experimental collision cross sections (CCS) for the fullerene C80- and the endohedral metallofullerenes La2@C80-, Sc3N@C80-, and Er3N@C80- in molecular nitrogen. The CCS values of the endohedral fullerenes are 0.2% larger than that of the empty cage. Using a combination of density functional theory and trajectory calculations, we were able to reproduce these experimental findings theoretically. Two effects are discussed that contribute to the CCS differences: (i) a small increase in fullerene cage size upon endohedral doping and (ii) charge transfer from the encapsulated moieties to the cage thus increasing the attractive charge-induced dipole interaction between the (endohedral) fullerene ion and the nitrogen bath gas molecules.

Compositions and Isomer Separation of Palladium Oxide Cluster Cations Studied by Ion Mobility Mass Spectrometry

Geometric structures of gas-phase palladium oxide cluster cations, PdnOm+, were investigated for stable compositions by ion mobility mass spectrometry (IMMS) and quantum chemical calculations. Pure metallic (m = 0) and oxygen-deficient (m < n) cluster cations were preferentially obtained from the mass spectra as a result of collision-induced dissociation. Structures of cluster series, Pd3Om+ (m = 1–6), Pd4Om+ (m = 2–8), and Pd5Om+ (m = 3–8), were determined by comparing experimental collision cross sections obtained by IMMS and theoretical collision cross sections of optimized structures by density functional theory calculations. As for the Pd3Om+ cluster cations, structural transition was observed from one-dimensional chains to two-dimensional (2D) branched/2D sheets and finally to three-dimensional (3D) compact structures with increasing m. These 2D and 3D isomers were found to retain their triangular metal-core configuration. 2D sheets and 3D compact isomers that maintain a tetrahedral metal-core confi...

ISiCLE: A Quantum Chemistry Pipeline for Establishing in Silico Collision Cross Section Libraries.

High-throughput, comprehensive, and confident identifications of metabolites and other chemicals in biological and environmental samples will revolutionize our understanding of the role these chemically diverse molecules play in biological systems. Despite recent technological advances, metabolomics studies still result in the detection of a disproportionate number of features that cannot be confidently assigned to a chemical structure. This inadequacy is driven by the single most significant limitation in metabolomics, the reliance on reference libraries constructed by analysis of authentic reference materials with limited commercial availability. To this end, we have developed the in silico chemical library engine (ISiCLE), a high-performance computing-friendly cheminformatics workflow for generating libraries of chemical properties. In the instantiation described here, we predict probable three-dimensional molecular conformers (i.e., conformational isomers) using chemical identifiers as input, from which collision cross sections (CCS) are derived. The approach employs first-principles simulation, distinguished by the use of molecular dynamics, quantum chemistry, and ion mobility calculations, to generate structures and chemical property libraries, all without training data. Importantly, optimization of ISiCLE included a refactoring of the popular MOBCAL code for trajectory-based mobility calculations, improving its computational efficiency by over 2 orders of magnitude. Calculated CCS values were validated against 1983 experimentally measured CCS values and compared to previously reported CCS calculation approaches. Average calculated CCS error for the validation set is 3.2% using standard parameters, outperforming other density functional theory (DFT)-based methods and machine learning methods (e.g., MetCCS). An online database is introduced for sharing both calculated and experimental CCS values ( metabolomics.pnnl.gov ), initially including a CCS library with over 1 million entries. Finally, three successful applications of molecule characterization using calculated CCS are described, including providing evidence for the presence of an environmental degradation product, the separation of molecular isomers, and an initial characterization of complex blinded mixtures of exposure chemicals. This work represents a method to address the limitations of small molecule identification and offers an alternative to generating chemical identification libraries experimentally by analyzing authentic reference materials. All code is available at github.com/pnnl .