Spatial lipidomics maps brain alterations associated with mild traumatic brain injury.

Traumatic brain injury (TBI) is a global public health problem with 50-60 million incidents per year, most of which are considered mild (mTBI) and many of these repetitive (rmTBI). Despite their massive implications, the pathologies of mTBI and rmTBI are not fully understood, with a paucity of information on brain lipid dysregulation following mild injury event(s). To gain more insight on mTBI and rmTBI pathology, a non-targeted spatial lipidomics workflow utilizing high resolution mass spectrometry imaging was developed to map brain region-specific lipid alterations in rats following injury. Discriminant multivariate models were created for regions of interest including the hippocampus, cortex, and corpus callosum to pinpoint lipid species that differentiated between injured and sham animals. A multivariate model focused on the hippocampus region differentiated injured brain tissues with an area under the curve of 0.99 using only four lipid species. Lipid classes that were consistently discriminant included polyunsaturated fatty acid-containing phosphatidylcholines (PC), lysophosphatidylcholines (LPC), LPC-plasmalogens (LPC-P) and PC potassium adducts. Many of the polyunsaturated fatty acid-containing PC and LPC-P selected have never been previously reported as altered in mTBI. The observed lipid alterations indicate that neuroinflammation and oxidative stress are important pathologies that could serve to explain cognitive deficits associated with rmTBI. Therapeutics which target or attenuate these pathologies may be beneficial to limit persistent damage following a mild brain injury event.

Sexual dimorphism and the multi-omic response to exercise training in rat subcutaneous white adipose tissue.

Subcutaneous white adipose tissue (scWAT) is a dynamic storage and secretory organ that regulates systemic homeostasis, yet the impact of endurance exercise training (ExT) and sex on its molecular landscape is not fully established. Utilizing an integrative multi-omics approach, and leveraging data generated by the Molecular Transducers of Physical Activity Consortium (MoTrPAC), we show profound sexual dimorphism in the scWAT of sedentary rats and in the dynamic response of this tissue to ExT. Specifically, the scWAT of sedentary females displays -omic signatures related to insulin signaling and adipogenesis, whereas the scWAT of sedentary males is enriched in terms related to aerobic metabolism. These sex-specific -omic signatures are preserved or amplified with ExT. Integration of multi-omic analyses with phenotypic measures identifies molecular hubs predicted to drive sexually distinct responses to training. Overall, this study underscores the powerful impact of sex on adipose tissue biology and provides a rich resource to investigate the scWAT response to ExT.

Automated Machine Learning and Explainable AI (AutoML-XAI) for Metabolomics: Improving Cancer Diagnostics.

Metabolomics generates complex data necessitating advanced computational methods for generating biological insight. While machine learning (ML) is promising, the challenges of selecting the best algorithms and tuning hyperparameters, particularly for nonexperts, remain. Automated machine learning (AutoML) can streamline this process; however, the issue of interpretability could persist. This research introduces a unified pipeline that combines AutoML with explainable AI (XAI) techniques to optimize metabolomics analysis. We tested our approach on two data sets: renal cell carcinoma (RCC) urine metabolomics and ovarian cancer (OC) serum metabolomics. AutoML, using Auto-sklearn, surpassed standalone ML algorithms like SVM and k-Nearest Neighbors in differentiating between RCC and healthy controls, as well as OC patients and those with other gynecological cancers. The effectiveness of Auto-sklearn is highlighted by its AUC scores of 0.97 for RCC and 0.85 for OC, obtained from the unseen test sets. Importantly, on most of the metrics considered, Auto-sklearn demonstrated a better classification performance, leveraging a mix of algorithms and ensemble techniques. Shapley Additive Explanations (SHAP) provided a global ranking of feature importance, identifying dibutylamine and ganglioside GM(d34:1) as the top discriminative metabolites for RCC and OC, respectively. Waterfall plots offered local explanations by illustrating the influence of each metabolite on individual predictions. Dependence plots spotlighted metabolite interactions, such as the connection between hippuric acid and one of its derivatives in RCC, and between GM3(d34:1) and GM3(18:1_16:0) in OC, hinting at potential mechanistic relationships. Through decision plots, a detailed error analysis was conducted, contrasting feature importance for correctly versus incorrectly classified samples. In essence, our pipeline emphasizes the importance of harmonizing AutoML and XAI, facilitating both simplified ML application and improved interpretability in metabolomics data science.

The mitochondrial multi-omic response to exercise training across rat tissues.

Mitochondria have diverse functions critical to whole-body metabolic homeostasis. Endurance training alters mitochondrial activity, but systematic characterization of these adaptations is lacking. Here, the Molecular Transducers of Physical Activity Consortium mapped the temporal, multi-omic changes in mitochondrial analytes across 19 tissues in male and female rats trained for 1, 2, 4, or 8 weeks. Training elicited substantial changes in the adrenal gland, brown adipose, colon, heart, and skeletal muscle. The colon showed non-linear response dynamics, whereas mitochondrial pathways were downregulated in brown adipose and adrenal tissues. Protein acetylation increased in the liver, with a shift in lipid metabolism, whereas oxidative proteins increased in striated muscles. Exercise-upregulated networks were downregulated in human diabetes and cirrhosis. Knockdown of the central network protein 17-beta-hydroxysteroid dehydrogenase 10 (HSD17B10) elevated oxygen consumption, indicative of metabolic stress. We provide a multi-omic, multi-tissue, temporal atlas of the mitochondrial response to exercise training and identify candidates linked to mitochondrial dysfunction.

Triboelectric Nanogenerators for the Masses: A Low-Cost Do-It-Yourself Pulsed Ion Source for Sample-Limited Applications.

Triboelectric nanogenerators (TENG) are useful devices for converting mechanical motion into electric current using readily available materials. Though the applications for these devices span across many fields, TENG can be leveraged for mass spectrometry (MS) as inexpensive and effective power supplies for pulsed nanoelectrospray ionization (nESI). The inherently discontinuous spray provided by TENG is particularly useful in scenarios where high sample economy is imperative, as in the case of ultraprecious samples. Previous work has shown the utility of TENG MS as a highly sensitive technique capable of yielding quality spectra from only a few microliters of sample at low micromolar concentrations. As the field of miniaturized, fieldable mass spectrometers grows, it remains critical to develop advanced ion sources with similarly small power requirements and footprints. Here, we present a redesigned TENG ion source with a sub-1000 USD material cost, lower power consumption, reduced footprint, and improved capabilities. We validate the performance of this new device for a diverse set of applications, including lipid double bond localization and native protein analysis.

Molecular Transducers of Physical Activity Consortium (MoTrPAC): Human Studies Design and Protocol.

Physical activity, including structured exercise, is associated with favorable health-related chronic disease outcomes. While there is evidence of various molecular pathways that affect these responses, a comprehensive molecular map of these molecular responses to exercise has not been developed. The Molecular Transducers of Physical Activity Consortium (MoTrPAC) is a multi-center study designed to isolate the effects of structured exercise training on the molecular mechanisms underlying the health benefits of exercise and physical activity. MoTrPAC contains both a pre-clinical and human component. The details of the human studies component of MoTrPAC that include the design and methods are presented here. The human studies contain both an adult and pediatric component. In the adult component, sedentary participants are randomized to 12 weeks of Control, Endurance Exercise Training, or Resistance Exercise Training with outcomes measures completed before and following the 12 weeks. The adult component also includes recruitment of highly active endurance trained or resistance trained participants who only complete measures once. A similar design is used for the pediatric component; however, only endurance exercise is examined. Phenotyping measures include weight, body composition, vital signs, cardiorespiratory fitness, muscular strength, physical activity and diet, and other questionnaires. Participants also complete an acute rest period (adults only) or exercise session (adults, pediatrics) with collection of biospecimens (blood only for pediatrics) to allow for examination of the molecular responses. The design and methods of MoTrPAC may inform other studies. Moreover, MoTrPAC will provide a repository of data that can be used broadly across the scientific community.

Integrative Analysis of Cytokine and Lipidomics Datasets Following Mild Traumatic Brain Injury in Rats.

Traumatic brain injury (TBI) is a significant source of disability in the United States and around the world and may lead to long-lasting cognitive deficits and a decreased quality of life for patients across injury severities. Following the primary injury phase, TBI is characterized by complex secondary cascades that involve altered homeostasis and metabolism, faulty signaling, neuroinflammation, and lipid dysfunction. The objectives of the present study were to (1) assess potential correlations between lipidome and cytokine changes after closed-head mild TBI (mTBI), and (2) examine the reproducibility of our acute lipidomic profiles following TBI. Cortices from 54 Sprague Dawley male and female rats were analyzed by ultra-high-performance liquid chromatography mass spectrometry (LC-MS) in both positive and negative ionization modes and multiplex cytokine analysis after single (smTBI) or repetitive (rmTBI) closed-head impacts, or sham conditions. Tissue age was a variable, given that two cohorts (n = 26 and n = 28) were initially run a year-and-a-half apart, creating inter-batch variations. We annotated the lipidome datasets using an in-house data dictionary based on exact masses of precursor and fragment ions and removed features with statistically significant differences between sham control batches. Our results indicate that lipids with high-fold change between injury groups moderately correlate with the cytokines eotaxin, IP-10, and TNF-α. Additionally, we show a significant decrease in the pro-inflammatory markers IL-1ß and IP-10, TNF-α, and RANTES in the rmTBI samples relative to the sham control. We discuss the major challenges in correlating high dimensional lipidomic data with functional cytokine profiles and the implications for understanding the biological significance of two related but disparate analysis modes in the study of TBI, an inherently heterogeneous neurological disorder.

metabCombiner 2.0: Disparate Multi-Dataset Feature Alignment for LC-MS Metabolomics.

Liquid chromatography-high-resolution mass spectrometry (LC-HRMS), as applied to untargeted metabolomics, enables the simultaneous detection of thousands of small molecules, generating complex datasets. Alignment is a crucial step in data processing pipelines, whereby LC-MS features derived from common ions are assembled into a unified matrix amenable to further analysis. Variability in the analytical factors that influence liquid chromatography separations complicates data alignment. This is prominent when aligning data acquired in different laboratories, generated using non-identical instruments, or between batches from large-scale studies. Previously, we developed metabCombiner for aligning disparately acquired LC-MS metabolomics datasets. Here, we report significant upgrades to metabCombiner that enable the stepwise alignment of multiple untargeted LC-MS metabolomics datasets, facilitating inter-laboratory reproducibility studies. To accomplish this, a "primary" feature list is used as a template for matching compounds in "target" feature lists. We demonstrate this workflow by aligning four lipidomics datasets from core laboratories generated using each institution's in-house LC-MS instrumentation and methods. We also introduce batchCombine, an application of the metabCombiner framework for aligning experiments composed of multiple batches. metabCombiner is available as an R package on Github and Bioconductor, along with a new online version implemented as an R Shiny App.

Serum Lipidome Profiling Reveals a Distinct Signature of Ovarian Cancer in Korean Women.

BACKGROUND: Distinguishing ovarian cancer from other gynecological malignancies is crucial for patient survival yet hindered by non-specific symptoms and limited understanding of ovarian cancer pathogenesis. Accumulating evidence suggests a link between ovarian cancer and deregulated lipid metabolism. Most studies have small sample sizes, especially for early-stage cases, and lack racial/ethnic diversity, necessitating more inclusive research for improved ovarian cancer diagnosis and prevention. METHODS: Here, we profiled the serum lipidome of 208 ovarian cancer, including 93 early-stage patients with ovarian cancer and 117 nonovarian cancer (other gynecological malignancies) patients of Korean descent. Serum samples were analyzed with a high-coverage liquid chromatography high-resolution mass spectrometry platform, and lipidome alterations were investigated via statistical and machine learning (ML) approaches. RESULTS: We found that lipidome alterations unique to ovarian cancer were present in Korean women as early as when the cancer is localized, and those changes increase in magnitude as the diseases progresses. Analysis of relative lipid abundances revealed specific patterns for various lipid classes, with most classes showing decreased abundance in ovarian cancer in comparison with other gynecological diseases. ML methods selected a panel of 17 lipids that discriminated ovarian cancer from nonovarian cancer cases with an AUC value of 0.85 for an independent test set. CONCLUSIONS: This study provides a systemic analysis of lipidome alterations in human ovarian cancer, specifically in Korean women. IMPACT: Here, we show the potential of circulating lipids in distinguishing ovarian cancer from nonovarian cancer conditions.

Spatial Lipidomics Maps Brain Alterations Associated with Mild Traumatic Brain Injury.

Traumatic brain injury (TBI) is a global public health problem with 50-60 million incidents per year, most of which are considered mild (mTBI) and many of these repetitive (rmTBI). Despite their massive implications, the pathologies of mTBI and rmTBI are not fully understood, with a paucity of information on brain lipid dysregulation following mild injury event(s). To gain more insight on mTBI and rmTBI pathology, a non-targeted spatial lipidomics workflow utilizing ultrahigh resolution mass spectrometry imaging was developed to map brain region-specific lipid alterations in rats following injury. Discriminant multivariate models were created for regions of interest including the hippocampus, cortex, and corpus callosum to pinpoint lipid species that differentiated between injured and sham animals. A multivariate model focused on the hippocampus region differentiated injured brain tissues with an area under the curve of 0.994 using only four lipid species. Lipid classes that were consistently discriminant included polyunsaturated fatty acid-containing phosphatidylcholines (PC), lysophosphatidylcholines (LPC), LPC-plasmalogens (LPC-P) and PC potassium adducts. Many of the polyunsaturated fatty acid-containing PC and LPC-P selected have never been previously reported as altered in mTBI. The observed lipid alterations indicate that neuroinflammation, oxidative stress and disrupted sodium-potassium pumps are important pathologies that could serve to explain cognitive deficits associated with rmTBI. Therapeutics which target or attenuate these pathologies may be beneficial to limit persistent damage following a mild brain injury event.

Comparing Brain and Blood Lipidome Changes following Single and Repetitive Mild Traumatic Brain Injury in Rats.

Traumatic brain injury (TBI) is a major health concern in the United States and globally, contributing to disability and long-term neurological problems. Lipid dysregulation after TBI is underexplored, and a better understanding of lipid turnover and degradation could point to novel biomarker candidates and therapeutic targets. Here, we investigated overlapping lipidome changes in the brain and blood using a data-driven discovery approach to understand lipid alterations in the brain and serum compartments acutely following mild TBI (mTBI) and the potential efflux of brain lipids to peripheral blood. The cortices and sera from male and female Sprague-Dawley rats were analyzed via ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS) in both positive and negative ion modes following single and repetitive closed head impacts. The overlapping lipids in the data sets were identified with an in-house data dictionary for investigating lipid class changes. MS-based lipid profiling revealed overall increased changes in the serum compartment, while the brain lipids primarily showed decreased changes. Interestingly, there were prominent alterations in the sphingolipid class in the brain and blood compartments after single and repetitive injury, which may suggest efflux of brain sphingolipids into the blood after TBI. Genetic algorithms were used for predictive panel selection to classify injured and control samples with high sensitivity and specificity. These overlapping lipid panels primarily mapped to the glycerophospholipid metabolism pathway with Benjamini-Hochberg adjusted q-values less than 0.05. Collectively, these results detail overlapping lipidome changes following mTBI in the brain and blood compartments, increasing our understanding of TBI-related lipid dysregulation while identifying novel biomarker candidates.

Divergent Skeletal Muscle Metabolomic Signatures of Different Exercise Training Modes Independently Predict Cardiometabolic Risk Factors.

We investigated the link between enhancement of SI (by hyperinsulinemic-euglycemic clamp) and muscle metabolites after 12 weeks of aerobic (high-intensity interval training [HIIT]), resistance training (RT), or combined training (CT) exercise in 52 lean healthy individuals. Muscle RNA sequencing revealed a significant association between SI after both HIIT and RT and the branched-chain amino acid (BCAA) metabolic pathway. Concurrently with increased expression and activity of branched-chain ketoacid dehydrogenase enzyme, many muscle amino metabolites, including BCAAs, glutamate, phenylalanine, aspartate, asparagine, methionine, and γ-aminobutyric acid, increased with HIIT, supporting the substantial impact of HIIT on amino acid metabolism. Short-chain C3 and C5 acylcarnitines were reduced in muscle with all three training modes, but unlike RT, both HIIT and CT increased tricarboxylic acid metabolites and cardiolipins, supporting greater mitochondrial activity with aerobic training. Conversely, RT and CT increased more plasma membrane phospholipids than HIIT, suggesting a resistance exercise effect on cellular membrane protection against environmental damage. Sex and age contributed modestly to the exercise-induced changes in metabolites and their association with cardiometabolic parameters. Integrated transcriptomic and metabolomic analyses suggest various clusters of genes and metabolites are involved in distinct effects of HIIT, RT, and CT. These distinct metabolic signatures of different exercise modes independently link each type of exercise training to improved SI and cardiometabolic risk. ARTICLE HIGHLIGHTS: We aimed to understand the link between skeletal muscle metabolites and cardiometabolic health after exercise training. Although aerobic, resistance, and combined exercise training each enhance muscle insulin sensitivity as well as other cardiometabolic parameters, they disparately alter amino and citric acid metabolites as well as the lipidome, linking these metabolomic changes independently to the improvement of cardiometabolic risks with each exercise training mode. These findings reveal an important layer of the unique exercise mode-dependent changes in muscle metabolism, which may eventually lead to more informed exercise prescription for improving SI.

Triboelectric Nanogenerators: Low-Cost Power Supplies for Improved Electrospray Ionization.

Electrospray ionization (ESI) is one of the most popular methods to generate ions for mass spectrometry (MS). When compared with other ionization techniques, it can generate ions from liquid-phase samples without additives, retaining covalent and non-covalent interactions of the molecules of interest. When hyphenated to liquid chromatography, it greatly expands the versatility of MS analysis of complex mixtures. However, despite the extensive growth in the application of ESI, the technique still suffers from some drawbacks when powered by direct current (DC) power supplies. Triboelectric nanogenerators promise to be a new power source for the generation of ions by ESI, improving on the analytical capabilities of traditional DC ESI. In this review we highlight the fundamentals of ESI driven by DC power supplies, its contrasting qualities to triboelectric nanogenerator power supplies, and its applications to three distinct fields of research: forensics, metabolomics, and protein structure analysis.

Automated machine learning and explainable AI (AutoML-XAI) for metabolomics: improving cancer diagnostics.

Motivation: Metabolomics generates complex data necessitating advanced computational methods for generating biological insight. While machine learning (ML) is promising, the challenges of selecting the best algorithms and tuning hyperparameters, particularly for non-experts, remain. Automated machine learning (AutoML) can streamline this process; however, the issue of interpretability could persist. This research introduces a unified pipeline that combines AutoML with explainable AI (XAI) techniques to optimize metabolomics analysis. Results: We tested our approach on two datasets: renal cell carcinoma (RCC) urine metabolomics and ovarian cancer (OC) serum metabolomics. AutoML, using auto-sklearn, surpassed standalone ML algorithms such as SVM and random forest in differentiating between RCC and healthy controls, as well as OC patients and those with other gynecological cancers (Non-OC). Auto-sklearn employed a mix of algorithms and ensemble techniques, yielding a superior performance (AUC of 0.97 for RCC and 0.85 for OC). Shapley Additive Explanations (SHAP) provided a global ranking of feature importance, identifying dibutylamine and ganglioside GM(d34:1) as the top discriminative metabolites for RCC and OC, respectively. Waterfall plots offered local explanations by illustrating the influence of each metabolite on individual predictions. Dependence plots spotlighted metabolite interactions, such as the connection between hippuric acid and one of its derivatives in RCC, and between GM3(d34:1) and GM3(18:1_16:0) in OC, hinting at potential mechanistic relationships. Through decision plots, a detailed error analysis was conducted, contrasting feature importance for correctly versus incorrectly classified samples. In essence, our pipeline emphasizes the importance of harmonizing AutoML and XAI, facilitating both simplified ML application and improved interpretability in metabolomics data science. Availability: https://github.com/obifarin/automl-xai-metabolomics.

Ultrahigh Resolution Lipid Mass Spectrometry Imaging of High-Grade Serous Ovarian Cancer Mouse Models.

No effective screening tools for ovarian cancer (OC) exist, making it one of the deadliest cancers among women. Considering little is known about the detailed progression and metastasis mechanism of OC at a molecular level, it is crucial to gain more insights on how metabolic and signaling alterations accompany its development. Herein, we present a comprehensive study using ultra-high-resolution Fourier transform ion cyclotron resonance matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI) to investigate the spatial distribution and alterations of lipids in ovarian tissues collected from double knockout (n = 4) and a triple mutant mouse models (n = 4) of high-grade serous ovarian cancer (HGSC). Lipids belonging to a total of 15 different classes were annotated and their abundance changes compared to those in healthy mouse reproductive tissue (n = 4), mapping onto major lipid pathways involved in OC progression. From intermediate-stage OC to advanced HGSC, we provide a direct visualization of lipid distributions and their biological links to inflammatory response, cellular stress, cell proliferation, and other processes. We also show the ability to distinguish tumors at different stages from healthy tissues via a number of highly specific lipid biomarkers, providing targets for future panels that could be useful in diagnosis.

Development of a Robust Consensus Modeling Approach for Identifying Cellular and Media Metabolites Predictive of Mesenchymal Stromal Cell Potency.

Mesenchymal stromal cells (MSCs) have shown promise in regenerative medicine applications due in part to their ability to modulate immune cells. However, MSCs demonstrate significant functional heterogeneity in terms of their immunomodulatory function because of differences in MSC donor/tissue source, as well as non-standardized manufacturing approaches. As MSC metabolism plays a critical role in their ability to expand to therapeutic numbers ex vivo, we comprehensively profiled intracellular and extracellular metabolites throughout the expansion process to identify predictors of immunomodulatory function (T-cell modulation and indoleamine-2,3-dehydrogenase (IDO) activity). Here, we profiled media metabolites in a non-destructive manner through daily sampling and nuclear magnetic resonance (NMR), as well as MSC intracellular metabolites at the end of expansion using mass spectrometry (MS). Using a robust consensus machine learning approach, we were able to identify panels of metabolites predictive of MSC immunomodulatory function for 10 independent MSC lines. This approach consisted of identifying metabolites in 2 or more machine learning models and then building consensus models based on these consensus metabolite panels. Consensus intracellular metabolites with high predictive value included multiple lipid classes (such as phosphatidylcholines, phosphatidylethanolamines, and sphingomyelins) while consensus media metabolites included proline, phenylalanine, and pyruvate. Pathway enrichment identified metabolic pathways significantly associated with MSC function such as sphingolipid signaling and metabolism, arginine and proline metabolism, and autophagy. Overall, this work establishes a generalizable framework for identifying consensus predictive metabolites that predict MSC function, as well as guiding future MSC manufacturing efforts through identification of high-potency MSC lines and metabolic engineering.

Machine Learning Reveals Lipidome Remodeling Dynamics in a Mouse Model of Ovarian Cancer.

Ovarian cancer (OC) is one of the deadliest cancers affecting the female reproductive system. It may present little or no symptoms at the early stages and typically unspecific symptoms at later stages. High-grade serous ovarian cancer (HGSC) is the subtype responsible for most ovarian cancer deaths. However, very little is known about the metabolic course of this disease, particularly in its early stages. In this longitudinal study, we examined the temporal course of serum lipidome changes using a robust HGSC mouse model and machine learning data analysis. Early progression of HGSC was marked by increased levels of phosphatidylcholines and phosphatidylethanolamines. In contrast, later stages featured more diverse lipid alterations, including fatty acids and their derivatives, triglycerides, ceramides, hexosylceramides, sphingomyelins, lysophosphatidylcholines, and phosphatidylinositols. These alterations underscored unique perturbations in cell membrane stability, proliferation, and survival during cancer development and progression, offering potential targets for early detection and prognosis of human ovarian cancer.

Mass Spectrometry Imaging Reveals Early Metabolic Priming of Cell Lineage in Differentiating Human-Induced Pluripotent Stem Cells.

Induced pluripotent stem cells (iPSCs) hold great promise in regenerative medicine; however, few algorithms of quality control at the earliest stages of differentiation have been established. Despite lipids having known functions in cell signaling, their role in pluripotency maintenance and lineage specification is underexplored. We investigated the changes in iPSC lipid profiles during the initial loss of pluripotency over the course of spontaneous differentiation using the co-registration of confocal microscopy and matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging. We identified phosphatidylethanolamine (PE) and phosphatidylinositol (PI) species that are highly informative of the temporal stage of differentiation and can reveal iPS cell lineage bifurcation occurring metabolically. Several PI species emerged from the machine learning analysis of MS data as the early metabolic markers of pluripotency loss, preceding changes in the pluripotency transcription factor Oct4. The manipulation of phospholipids via PI 3-kinase inhibition during differentiation manifested in the spatial reorganization of the iPS cell colony and elevated expression of NCAM-1. In addition, the continuous inhibition of phosphatidylethanolamine N-methyltransferase during differentiation resulted in the enhanced maintenance of pluripotency. Our machine learning analysis highlights the predictive power of lipidomic metrics for evaluating the early lineage specification in the initial stages of spontaneous iPSC differentiation.