Biological Insights Research Articles

Robustness and reliability of single-cell regulatory multi-omics with deep mitochondrial mutation profiling.

The detection of mitochondrial DNA (mtDNA) mutations in single cells holds considerable potential to define clonal relationships coupled with information on cell state in humans. Previous methods focused on higher heteroplasmy mutations that are limited in number and can be influenced by functional selection, introducing biases for lineage tracing. Although more challenging to detect, intermediate to low heteroplasmy mtDNA mutations are valuable due to their high diversity, abundance, and lower propensity to selection. To enhance mtDNA mutation detection and facilitate fine-scale lineage tracing, we developed the single-cell Regulatory multi-omics with Deep Mitochondrial mutation profiling (ReDeeM) approach, an integrated experimental and computational framework. Recently, some concerns have been raised about the analytical workflow in the ReDeeM framework. Specifically, it was noted that the mutations detected in a single molecule per cell are enriched on edges of mtDNA molecules, suggesting they resemble artifacts reported in other sequencing approaches. It was then proposed that all mutations found in one molecule per cell should be removed. We detail our error correction method, demonstrating that the observed edge mutations are distinct from previously reported sequencing artifacts. We further show that the proposed removal leads to massive elimination of bona fide and informative mutations. Indeed, mutations accumulating on edges impact a minority of all mutation calls (for example, in hematopoietic stem cells, the excess mutations on the edge account for only 4.3%-7.6% of the total). Recognizing the value of addressing edge mutations even after applying consensus correction, we provide an additional filtering option in the ReDeeM-R package. This approach effectively eliminates the position biases, leads to a mutational signature indistinguishable from bona fide mitochondrial mutations, and removes excess low molecule high connectedness mutations. Importantly, this option preserves the large majority of unique mutations identified by ReDeeM, maintaining the ability of ReDeeM to provide a more than 10-fold increase in variant detection compared to previous methods. Additionally, the cells remain well-connected. While there is room for further refinement in mutation calling strategies, the significant advances and biological insights provided by the ReDeeM framework are unique and remain intact. We hope that this detailed discussion and analysis enables the community to employ this approach and contribute to its further development.

GRable Version 1.0: A Software Tool for Site-Specific Glycoform Analysis With Improved MS1-Based Glycopeptide Detection With Parallel Clustering and Confidence Evaluation With MS2 Information

High-throughput intact glycopeptide analysis is crucial for elucidating the physiological and pathological status of the glycans attached to each glycoprotein. Mass spectrometry-based glycoproteomic methods are challenging because of the diversity and heterogeneity of glycan structures. Therefore, we developed an MS1-based site-specific glycoform analysis method named "Glycan heterogeneity-based Relational IDentification of Glycopeptide signals on Elution profile (Glyco-RIDGE)" for a more comprehensive analysis. This method detects glycopeptide signals as a cluster based on the mass and chromatographic properties of glycopeptides and then searches for each combination of core peptides and glycan compositions by matching their mass and retention time differences. Here, we developed a novel browser-based software named GRable for semi-automated Glyco-RIDGE analysis with significant improvements in glycopeptide detection algorithms, including "parallel clustering." This unique function improved the comprehensiveness of glycopeptide detection and allowed the analysis to focus on specific glycan structures, such as pauci-mannose. The other notable improvement is evaluating the "confidence level" of the GRable results, especially using MS2 information. This function facilitated reduced misassignment of the core peptide and glycan composition and improved the interpretation of the results. Additional improved points of the algorithms are "correction function" for accurate monoisotopic peak picking; one-to-one correspondence of clusters and core peptides even for multiply sialylated glycopeptides; and "inter-cluster analysis" function for understanding the reason for detected but unmatched clusters. The significance of these improvements was demonstrated using purified and crude glycoprotein samples, showing that GRable allowed site-specific glycoform analysis of intact sialylated glycoproteins on a large-scale and in-depth. Therefore, this software will help us analyze the status and changes in glycans to obtain biological and clinical insights into protein glycosylation by complementing the comprehensiveness of MS2-based glycoproteomics. GRable can be freely run online using a web browser via the GlyCosmos Portal (https://glycosmos.org/grable).

MetDecode: methylation-based deconvolution of cell-free DNA for noninvasive multi-cancer typing.

Circulating-cell free DNA (cfDNA) is widely explored as a noninvasive biomarker for cancer screening and diagnosis. The ability to decode the cells of origin in cfDNA would provide biological insights into pathophysiological mechanisms, aiding in cancer characterization and directing clinical management and follow-up. We developed a DNA methylation signature-based deconvolution algorithm, MetDecode, for cancer tissue origin identification. We built a reference atlas exploiting de novo and published whole-genome methylation sequencing data for colorectal, breast, ovarian, and cervical cancer, and blood-cell-derived entities. MetDecode models the contributors absent in the atlas with methylation patterns learnt on-the-fly from the input cfDNA methylation profiles. In addition, our model accounts for the coverage of each marker region to alleviate potential sources of noise. In-silico experiments showed a limit of detection down to 2.88% of tumor tissue contribution in cfDNA. MetDecode produced Pearson correlation coefficients above 0.95 and outperformed other methods in simulations (P < 0.001; T-test; one-sided). In plasma cfDNA profiles from cancer patients, MetDecode assigned the correct tissue-of-origin in 84.2% of cases. In conclusion, MetDecode can unravel alterations in the cfDNA pool components by accurately estimating the contribution of multiple tissues, while supplied with an imperfect reference atlas. MetDecode is available at https://github.com/JorisVermeeschLab/MetDecode.

Integrating mechanism-based T cell phenotypes into a model of tumor-immune cell interactions.

Interactions between cancer cells and immune cells in the tumor microenvironment influence tumor growth and can contribute to the response to cancer immunotherapies. It is difficult to gain mechanistic insights into the effects of cell-cell interactions in tumors using a purely experimental approach. However, computational modeling enables quantitative investigation of the tumor microenvironment, and agent-based modeling, in particular, provides relevant biological insights into the spatial and temporal evolution of tumors. Here, we develop a novel agent-based model (ABM) to predict the consequences of intercellular interactions. Furthermore, we leverage our prior work that predicts the transitions of CD8+ T cells from a naïve state to a terminally differentiated state using Boolean modeling. Given the details incorporated to predict T cell state, we apply the integrated Boolean-ABM framework to study how the properties of CD8+ T cells influence the composition and spatial organization of tumors and the efficacy of an immune checkpoint blockade. Overall, we present a mechanistic understanding of tumor evolution that can be leveraged to study targeted immunotherapeutic strategies.

Genome-wide study identifies novel genes associated with bone toxicities in children with acute lymphoblastic leukaemia.

Bone toxicities are common among paediatric patients treated for acute lymphoblastic leukaemia (ALL) with potentially major negative impact on patients' quality of life. To identify the underlying genetic contributors, we conducted a genome-wide association study (GWAS) and a transcriptome-wide association study (TWAS) in 260 patients of European-descent from the DFCI 05-001 ALL trial, with validation in 101 patients of European-descent from the DFCI 11-001 ALL trial. We identified a significant association between rs844882 on chromosome 20 and bone toxicities in the DFCI 05-001 trial (p = 1.7 × 10-8). In DFCI 11-001 trial, we observed a consistent trend of this variant with fracture. The variant was an eQTL for two nearby genes, CD93 and THBD. In TWAS, genetically predicted ACAD9 expression was associated with an increased risk of bone toxicities, which was confirmed by meta-analysis of the two cohorts (meta-p = 2.4 × 10-6). In addition, a polygenic risk score of heel quantitative ultrasound speed of sound was associated with fracture risk in both cohorts (meta-p = 2.3 × 10-3). Our findings highlight the genetic influence on treatment-related bone toxicities in this patient population. The genes we identified in our study provide new biological insights into the development of bone adverse events related to ALL treatment.

Constrained Pseudo-time Ordering for Clinical Transcriptomics Data.

Time series RNASeq studies can enable understanding of the dynamics of disease progression and treatment response in patients. They also provide information on biomarkers, activated and repressed pathways, and more. While useful, data from multiple patients is challenging to integrate due to the heterogeneity in treatment response among patients, and the small number of timepoints that are usually profiled. Due to the heterogeneity among patients, relying on the sampled time points to integrate data across individuals is challenging and does not lead to correct reconstruction of the response patterns. To address these challenges, we developed a new constrained based pseudotime ordering method for analyzing transcriptomics data in clinical and response studies. Our method allows the assignment of samples to their correct placement on the response curve while respecting the individual patient order. We use polynomials to represent gene expression over the duration of the study and an EM algorithm to determine parameters and locations. Application to three treatment response datasets shows that our method improves on prior methods and leads to accurate orderings that provide new biological insight on the disease and response. Code for the method is available at https://github.com/Sanofi-Public/ RDCS-bulkRNASeq-pseudo ordering.

LBFextract: Unveiling transcription factor dynamics from liquid biopsy data

MotivationThe analysis of circulating cell-free DNA (cfDNA) holds immense promise as a non-invasive diagnostic tool across various human conditions. However, extracting biological insights from cfDNA fragments entails navigating complex and diverse bioinformatics methods, encompassing not only DNA sequence variation, but also epigenetic characteristics like nucleosome footprints, fragment length, and methylation patterns. ResultsWe introduce Liquid Biopsy Feature extract (LBFextract), a comprehensive package designed to streamline feature extraction from cfDNA sequencing data, with the aim of enhancing the reproducibility and comparability of liquid biopsy studies. LBFextract facilitates the integration of preprocessing and postprocessing steps through alignment fragment tags and a hook mechanism. It incorporates various methods, including coverage-based and fragment length-based approaches, alongside two novel feature extraction methods: an entropy-based method to infer TF activity from fragmentomics data and a technique to amplify signals from nucleosome dyads. Additionally, it implements a method to extract condition-specific differentially active TFs based on these features for biomarker discovery. We demonstrate the use of LBFextract for the subtype classification of advanced prostate cancer patients using coverage signals at transcription factor binding sites from cfDNA. We show that LBFextract can generate robust and interpretable features that can discriminate between different clinical groups. LBFextract is a versatile and user-friendly package that can facilitate the analysis and interpretation of liquid biopsy data. Data and Code Availability and ImplementationLBFextract is freely accessible at https://github.com/Isy89/LBF. It is implemented in Python and compatible with Linux and Mac operating systems. Code and data to reproduce these analyses have been uploaded to 10.5281/zenodo.10964406.

Hypoxia-responsive synthetic gene circuits to improve safety and potency of CAR T cell therapy for solid tumors.

37 Background: Engineered cell therapies expressing a chimeric antigen receptor (CAR) in human T cells have demonstrated great promise for the treatment of hematological malignancies. However, unique challenges remain in translating the success of CAR T cell therapies to solid tumors, namely: (1) specificity: there are few truly unique antigens present on solid tumors, leading to off-target effects that damage normal tissue and pose safety risks; and (2) suppression: solid tumors develop microenvironments that are immunosuppressive and decrease T cell survival and tumor-directed cytotoxicity. To address these challenges and improve the safety and potency of solid tumor cell therapy, we engineered cells with synthetic genetic circuits containing a previously validated hypoxia biosensor (HBS) that responds to upregulation of the native HIF1α and HIF2α transcription factors by hypoxia, enabling the cells to sense and respond to the hypoxic tumor microenvironment ubiquitous among solid tumors. Methods: We evaluated CAR expression from circuit architectures containing the HBS promoter driving expression of both CAR and native HIF1α, HIF2α, as well as from circuit architectures containing the HBS promoter and our toolkit of synthetic transcription factors and promoters. Circuits were integrated into HEK293 cells via PiggyBac transposon vector to generate stable cell lines. Cells were subsequently cultured in normoxia (21% O2) and hypoxia (1% O2), and harvested for flow cytometric analysis to evaluate CAR surface expression across a three day time frame. Results: We identified novel circuits providing fast reporter output with minimal background signal under hypoxic conditions, and elucidated design rules that enable or restrict amplification of transgene in native HIF1α and HIF2α-containing circuits. We observed differences in CAR transgene expression in hypoxia between circuits containing HIF1α and HIF2α-mediated feedback, suggesting mechanistic differences in regulation of HIF1α and HIF2α when overexpressed. Additionally, we observed greatly amplified transgene expression in circuit topologies containing feedback mediated by our synthetic transcription factor and promoter toolkit compared to native HIF1α and HIF2α-containing circuits and cells constitutively expressing CAR. Conclusions: We present a new therapeutic technology that senses hypoxic environments and enables high-output sense and respond behaviors with therapeutic outputs. We envision that this technology will enable development of safer, more effective cell therapies for solid tumors, as well as provide new biological and mechanistic insights that can inform the future development of hypoxia-responsive cell therapies.

ExpOmics: a comprehensive web platform empowering biologists with robust multi-omics data analysis capabilities.

High-throughput technologies yield a broad spectrum of multi-omics datasets, which offer unparalleled insights into complex biological systems. However, effectively analyzing this diverse array of data presents challenges, considering factors such as species diversity, data types, costs, and limitations of the available tools. Herein, we present ExpOmics, a comprehensive web platform featuring 7 applications and 4 toolkits, with 28 customizable analysis functions spanning various analyses of differential expression, co-expression, Weighted Gene Co-expression Network Analysis (WGCNA), feature selection, and functional enrichment. ExpOmics allows users to upload and explore multi-omics data without organism restrictions, supporting various expression data, including genes, mRNAs, lncRNAs, miRNAs, circRNAs, piRNAs, and proteins and is compatible with diverse gene nomenclatures and expression values. Moreover, ExpOmics enables users to analyze 22427 transcriptomic datasets of 196 cancer subtypes sourced from 63 projects of The Cancer Genome Atlas Program (TCGA) to identify cancer biomarkers. The analysis results from ExpOmics are presented in high-quality graphical formats suitable for publication and are available for free download. A case study using ExpOmics identified two potential oncogenes, SERPINE1 and SLC43A1, that may regulate colorectal cancer through distinct biological processes. In summary, ExpOmics can serves as a robust platform for global researchers to explore multi-omics data, gain biological insights, and formulate testable hypotheses. ExpOmics is available at http://www.biomedical-web.com/expomics.

Identification of novel cattle (Bos taurus) genes and biological insights of their function in pre-implantation embryo development

BackgroundAppropriate regulation of genes expressed in oocytes and embryos is essential for acquisition of developmental competence in mammals. Here, we hypothesized that several genes expressed in oocytes and pre-implantation embryos remain unknown. Our goal was to reconstruct the transcriptome of oocytes (germinal vesicle and metaphase II) and pre-implantation cattle embryos (blastocysts) using short-read and long-read sequences to identify putative new genes.ResultsWe identified 274,342 transcript sequences and 3,033 of those loci do not match a gene present in official annotations and thus are potential new genes. Notably, 63.67% (1,931/3,033) of potential novel genes exhibited coding potential. Also noteworthy, 97.92% of the putative novel genes overlapped annotation with transposable elements. Comparative analysis of transcript abundance identified that 1,840 novel genes (recently added to the annotation) or potential new genes were differentially expressed between developmental stages (FDR < 0.01). We also determined that 522 novel or potential new genes (448 and 34, respectively) were upregulated at eight-cell embryos compared to oocytes (FDR < 0.01). In eight-cell embryos, 102 novel or putative new genes were co-expressed (|r|> 0.85, P < 1 × 10–8) with several genes annotated with gene ontology biological processes related to pluripotency maintenance and embryo development. CRISPR-Cas9 genome editing confirmed that the disruption of one of the novel genes highly expressed in eight-cell embryos reduced blastocyst development (ENSBTAG00000068261, P = 1.55 × 10–7).ConclusionsOur results revealed several putative new genes that need careful annotation. Many of the putative new genes have dynamic regulation during pre-implantation development and are important components of gene regulatory networks involved in pluripotency and blastocyst formation.

From Blur to Brilliance: The Ascendance of Advanced Microscopy in Neuronal Cell Biology.

The intricate network of the brain's neurons and synapses poses unparalleled challenges for research, distinct from other biological studies. This is particularly true when dissecting how neurons and their functional units work at a cell biological level. While traditional microscopy has been foundational, it was unable to reveal the deeper complexities of neural interactions. However, an imaging renaissance has transformed our capabilities. Advancements in light and electron microscopy, combined with correlative imaging, now achieve unprecedented resolutions, uncovering the most nuanced neural structures. Maximizing these tools requires more than just technical proficiency. It is crucial to align research aims, allocate resources wisely, and analyze data effectively. At the heart of this evolution is interdisciplinary collaboration, where various experts come together to translate detailed imagery into significant biological insights. This review navigates the latest developments in microscopy, underscoring both the promise of and prerequisites for bending this powerful tool set to understanding neuronal cell biology.

MOSBY enables multi-omic inference and spatial biomarker discovery from whole slide images

The utility of deep neural nets has been demonstrated for mapping hematoxylin-and-eosin (H&E) stained image features to expression of individual genes. However, these models have not been employed to discover clinically relevant spatial biomarkers. Here we develop MOSBY (Multi-Omic translation of whole slide images for Spatial Biomarker discoverY) that leverages contrastive self-supervised pretraining to extract improved H&E whole slide images features, learns a mapping between image and bulk omic profiles (RNA, DNA, and protein), and utilizes tile-level information to discover spatial biomarkers. We validate MOSBY gene and gene set predictions with spatial transcriptomic and serially-sectioned CD8 IHC image data. We demonstrate that MOSBY-inferred colocalization features have survival-predictive power orthogonal to gene expression, and enable concordance indices highly competitive with survival-trained multimodal networks. We identify and validate (1) an ER stress-associated colocalization feature as a chemotherapy-specific risk factor in lung adenocarcinoma, and (2) the colocalization of T effector cell vs cysteine signatures as a negative prognostic factor in multiple cancer indications. The discovery of clinically relevant biologically interpretable spatial biomarkers showcases the utility of the model in unraveling novel insights in cancer biology as well as informing clinical decision-making.

Pharma[e]cology: How the Gut Microbiome Contributes to Variations in Drug Response.

Drugs represent our first, and sometimes last, line of defense for many diseases, yet despite decades of research we still do not fully understand why a given drug works in one patient and fails in the next. The human gut microbiome is one of the missing puzzle pieces, due to its ability to parallel and extend host pathways for drug metabolism, along with more complex host-microbiome interactions. Herein, we focus on the well-established links between the gut microbiome and drugs for heart disease and cancer, plus emerging data on neurological disease. We highlight the interdisciplinary methods that are available and how they can be used to address major remaining knowledge gaps, including the consequences of microbial drug metabolism for treatment outcomes. Continued progress in this area promises fundamental biological insights into humans and their associated microbial communities and strategies for leveraging the microbiome to improve the practice of medicine.

Whole genome sequencing based analysis of inflammation biomarkers in the Trans-Omics for Precision Medicine (TOPMed) consortium.

Inflammation biomarkers can provide valuable insight into the role of inflammatory processes in many diseases and conditions. Sequencing based analyses of such biomarkers can also serve as an exemplar of the genetic architecture of quantitative traits. To evaluate the biological insight, which can be provided by a multi-ancestry, whole-genome based association study, we performed a comprehensive analysis of 21 inflammation biomarkers from up to 38 465 individuals with whole-genome sequencing from the Trans-Omics for Precision Medicine (TOPMed) program (with varying sample size by trait, where the minimum sample size was n = 737 for MMP-1). We identified 22 distinct single-variant associations across 6 traits-E-selectin, intercellular adhesion molecule 1, interleukin-6, lipoprotein-associated phospholipase A2 activity and mass, and P-selectin-that remained significant after conditioning on previously identified associations for these inflammatory biomarkers. We further expanded upon known biomarker associations by pairing the single-variant analysis with a rare variant set-based analysis that further identified 19 significant rare variant set-based associations with 5 traits. These signals were distinct from both significant single variant association signals within TOPMed and genetic signals observed in prior studies, demonstrating the complementary value of performing both single and rare variant analyses when analyzing quantitative traits. We also confirm several previously reported signals from semi-quantitative proteomics platforms. Many of these signals demonstrate the extensive allelic heterogeneity and ancestry-differentiated variant-trait associations common for inflammation biomarkers, a characteristic we hypothesize will be increasingly observed with well-powered, large-scale analyses of complex traits.

Genomic patterns of somatic mutations provide new prognostic, therapeutic, and biological insights in cancer

The mutational landscape of an individual's cancer can inform on its molecular state and be used as prognostic and therapeutic markers. The study by Barbour etal.1 analyzes mutational patterns in bladder cancer samples to uncover new biological insights into the ERCC2 gene function and develop new predictive prognostic tools.

Single-cell spatial transcriptomics reveals molecular patterns of selective neuronal vulnerability to α-synuclein pathology in a transgenic mouse model of Lewy body disease.

One of the unifying pathological hallmarks of Parkinson's disease (PD) and dementia with Lewy bodies (DLB) is the presence of misfolded, aggregated, and often phosphorylated forms of the protein α-synuclein in neurons. α-Synuclein pathology appears in select populations of neurons throughout various cortical and subcortical regions, and little is currently known about why some neurons develop pathology while others are spared. Here, we utilized subcellular-resolution imaging-based spatial transcriptomics (IST) in a transgenic mouse model that overexpresses wild-type human α-synuclein (α-syn-tg) to evaluate patterns of selective neuronal vulnerability to α-synuclein pathology. By performing post-IST immunofluorescence for α-synuclein phosphorylated at Ser129 (pSyn), we identified cell types in the cortex and hippocampus that were vulnerable or resistant to developing pSyn pathology. Next, we investigated the transcriptional underpinnings of the observed selective vulnerability using a set of custom probes to detect genes involved in α-synuclein processing and toxicity. We identified expression of the kinase:substrate pair Plk2 , which phosphorylates α-synuclein at Ser129, and human SNCA ( hSNCA ), as underlying the selective vulnerability to pSyn pathology. Finally, we performed differential gene expression analysis, comparing non-transgenic cells to pSyn - and pSyn + α-syn-tg cells to reveal gene expression changes downstream of hSNCA overexpression and pSyn pathology, which included pSyn-dependent alterations in mitochondrial and endolysosomal genes. This study provides a comprehensive use case of IST, yielding new biological insights into the formation of α-synuclein pathology and its downstream effects in a PD/DLB mouse model.

Discovery and generalization of tissue structures from spatial omics data

Tissues are organized into anatomical and functional units at different scales. New technologies for high-dimensional molecular profiling in situ have enabled the characterization of structure-function relationships in increasing molecular detail. However, it remains a challenge to consistently identify key functional units across experiments, tissues, and disease contexts, a task that demands extensive manual annotation. Here, we present spatial cellular graph partitioning (SCGP), a flexible method for the unsupervised annotation of tissue structures. We further present a reference-query extension pipeline, SCGP-Extension, that generalizes reference tissue structure labels to previously unseen samples, performing data integration and tissue structure discovery. Our experiments demonstrate reliable, robust partitioning of spatial data in a wide variety of contexts and best-in-class accuracy in identifying expertly annotated structures. Downstream analysis on SCGP-identified tissue structures reveals disease-relevant insights regarding diabetic kidney disease, skin disorder, and neoplastic diseases, underscoring its potential to drive biological insight and discovery from spatial datasets.

Applying interpretable machine learning in computational biology-pitfalls, recommendations and opportunities for new developments.

Recent advances in machine learning have enabled the development of next-generation predictive models for complex computational biology problems, thereby spurring the use of interpretable machine learning (IML) to unveil biological insights. However, guidelines for using IML in computational biology are generally underdeveloped. We provide an overview of IML methods and evaluation techniques and discuss common pitfalls encountered when applying IML methods to computational biology problems. We also highlight open questions, especially in the era of large language models, and call for collaboration between IML and computational biology researchers.

Interpreting and visualizing pathway analyses using embedding representations with PAVER.

Omics studies use large-scale high-throughput data to explain changes underlying different traits or conditions. However, omics analysis often results in long lists of pathways that are difficult to interpret. Therefore, it is of interest to describe a tool named PAVER (Pathway Analysis Visualization with Embedding Representations) for large scale genomic analysis. PAVER curates similar pathways into groups, identifies the pathway most representative of each group, and provides publication-ready intuitive visualizations. PAVER clusters pathways defined by their vector embedding representations and then identifies the term most cosine similar to its respective cluster's average embedding. PAVER can integrate multiple pathway analyses, highlight relevant biological insights, and work with any pathway database.

Neurometabolic topography and associations with cognition in Alzheimer's disease: A whole-brain high-resolution 3D MRSI study.

Altered neurometabolism, detectable via proton magnetic resonance spectroscopic imaging (1H-MRSI), is spatially heterogeneous and underpins cognitive impairments in Alzheimer's disease (AD). However, the spatial relationships between neurometabolic topography and cognitive impairment in AD remain unexplored due to technical limitations. We used a novel whole-brain high-resolution 1H-MRSI technique, with simultaneously acquired 18F-florbetapir positron emission tomography (PET) imaging, to investigate the relationship between neurometabolic topography and cognitive functions in 117 participants, including 22 prodromal AD, 51 AD dementia, and 44 controls. Prodromal AD and AD dementia patients exhibited spatially distinct reductions in N-acetylaspartate, and increases in myo-inositol. Reduced N-acetylaspartate and increased myo-inositol were associated with worse global cognitive performance, and N-acetylaspartate correlated with five specific cognitive scores. Neurometabolic topography provides biological insights into diverse cognitive dysfunctions. Whole-brain high-resolution 1H-MRSI revealed spatially distinct neurometabolic topographies associated with cognitive decline in AD, suggesting potential for noninvasive brain metabolic imaging to track AD progression. Whole-brain high-resolution 1H-MRSI unveils neurometabolic topography in AD. Spatially distinct reductions in NAA, and increases in mI, are demonstrated. NAA and mI topography correlates with global cognitive performance. NAA topography correlates with specific cognitive performance.