Biology subjects

Patel, R.

Publications and source records attributed to Patel, R..

At least 19 recordsLinked to original sources

bioRxiv2026

X-SPEAR: An Extensible Framework for Spatial Panel Evaluation and Refinement

Targeted spatial transcriptomics platforms such as Xenium require a predefined gene panel, and the genes in that panel are detected less efficiently, and to gene-specific degrees, in the spatial assay than in the single-cell reference used to choose them. We present X-SPEAR, a framework that couples DART marker selection with probe-level off-target assessment. DART learns short, interpretable Boolean gene rules and assembles several gene-disjoint rules per cell type. Its detection-aware training thins the reference to per-gene detection rates measured in a different, already completed experiment, so no Xenium data from the target tissue are needed before the panel is ordered. On held-out Xenium data from ovarian cancer, mouse brain and cervical cancer, DART substantially outperformed existing marker-selection methods at matched panel size, by 0.04-0.11 macro-F1 over Wilcoxon rank-sum, 0.04-0.08 over COSG and 0.10-0.18 over NS-Forest, and it reached Wilcoxon's best accuracy with 42-69% fewer genes. Ablations show that most of this gain comes from detection-aware training, which also improves other rank-based selectors and requires a detection profile measured with the same assay chemistry. At similar accuracy, DART's rule structure adds rules that faithfully summarise the downstream classifier, robustness to single-probe loss across 45 retinal ganglion cell subtypes, and better resolution of closely related cell states. We are applying X-SPEAR to design a Xenium panel for mouse retinal ganglion cell subtypes, and its in situ evaluation is in progress. X-SPEAR is available as an open-source Python package.

genomics↗

bioRxiv2026

Systematic exploration of predicted quaternary structures within pandemic-relevant viral proteomes

Mechanistic understanding of viral protein-protein interactions enables global health security and pandemic preparedness, yet experimental characterisation remains difficult, costly, and often restricted to specialised laboratories. Here, we describe an in silico campaign using AlphaFold2 and AlphaFold-Multimer to predict monomers and dimer structures within 2,812 viral proteomes from 23 viral families relevant to human health. We report high-confidence predicted structures of 5,279 hetero-, and 2,749 homo-dimers. Structural clustering of interfaces reduces the set fivefold to 1,598 consolidated by form and function, of which 471 (29.5%) lacked detectable similarity to experimentally determined interfaces in the PDB. These clusters yielded two notable findings: new data on vaccinology-relevant glycoproteins, and accurate prediction of the specificity and cleavage-site recognition of viral proteases, which are targets for anti-viral development. Viral monomers and high-confidence dimers are available for interactive browsing in the Pandemic Preparedness Portal of the AlphaFold Database (https://alphafold.ebi.ac.uk/), and the full dataset is available for bulk download.

bioinformatics↗

bioRxiv2026

Neofunctionalised plastid translocon regulates chloroplast ageing in Arabidopsis

Protein import across the chloroplast outer membrane is mediated by the TOC complex and regulated by GTP hydrolysis at two receptor GTPases, large and small. The former are the {beta}-barrel-containing Toc159 family of proteins (Sun and Jarvis 2023; H. Li and Chiu 2010; Shi and Theg 2013a; Kessler and Schnell 2009). The smallest member of this family in higher plants is Toc90, which has long been regarded as functionally unimportant due to the absence of obvious phenotypes in early mutant studies (Kubis et al. 2004; Infanger et al. 2011a). It was hypothesised that Toc90 is a relic of an ancestral Toc159 protein from early green lineage organisms (Hofmann and Theg 2003; Reumann et al. 2005). Here, we challenge this view by combining genetic, biochemical, phylogenetic, and in silico structural analyses to show that Toc90 is not an ancient remnant, but rather a neofunctionalised derivative of Toc159, and likely the most recently evolved member of the family. We show that Toc90 physically associates with Toc33 as a part of a photosynthetic TOC complex. Furthermore, we identify Toc90 as a regulator of chloroplast ageing. Our findings redefine Toc90 as an active and specialised component of the chloroplast protein import machinery, playing a pivotal role in senescence, especially in reducing photosynthetic protein abundance. Together, these results highlight Toc90 as a promising target for manipulating chloroplast longevity and plant senescence, with potential to improve crop productivity and yield.

plant biology↗

bioRxiv2026

Large language model-based bibliometric evaluation of population descriptors in human genetics

As the use of population descriptors such as race, ethnicity, and ancestry have become increasingly common in modern genetics research, there have been growing calls to critically examine their use. Most notably, in 2023, the National Academies of Science, Engineering, and Medicine (NASEM) published a report titled Using Population Descriptors in Genetics and Genomics Research: A New Framework for an Evolving Field, which included eight specific and actionable recommendations for researchers to implement the ethical and accurate use of population descriptors in genetic research. Here, we use the 2023 NASEM report as a benchmark to analyze the use of population descriptors in genome-wide association studies (GWAS). We develop a general toolkit for large language model-based bibliometrics, operationalize the report's recommendations into an evaluation framework, and apply this framework to evaluate all 4,007 papers from the GWAS Catalog published between 2007 and 2025 with full text available on PubMedCentral. We find significant improvements in adherence to NASEM report recommendations over time. However, most improvements predate the publication of the NASEM report itself, suggesting the report functioned primarily as a synthesis of existing best practices rather than a catalyst for change. We conclude by highlighting opportunities for growth in the field of human genetics.

genetics↗

bioRxiv2026

Potency of nucleoside analogs against Powassan virus replication and identification of 4'-fluorouridine as a therapeutic candidate for tick-borne orthoflavivirus encephalitis

Powassan virus (POWV) is an emerging tick-borne orthoflavivirus that causes severe encephalitis, long-term neurologic sequelae, and death, yet no approved antiviral therapies exist. Here, we evaluated several previously characterized nucleoside analogs as candidate countermeasures for POWV encephalitis. Favipiravir, molnupiravir, 4'-fluorouridine (4'-FlU), and GS-441524 inhibited POWV replication in vitro and demonstrated comparable activity against West Nile virus (WNV). We established C3H/HeNCrl mice as a stringent POWV disease model, showing increased susceptibility, higher day 7 brain titers, and more pronounced neurologic signs resembling severe human disease, compared with C57BL/6 mice. In prophylactic and therapeutic studies, 4'-fluorouridine was the most efficacious compound tested, significantly reducing brain viral burden even when treatment was delayed. Direct comparison with molnupiravir, favipiravir, and GS-441524 in mice confirmed superior survival benefit and antiviral activity for 4'-FlU against both POWV and WNV. These data identify 4'-fluorouridine as a promising therapeutic candidate for neurotropic flavivirus disease.

microbiology↗

bioRxiv2026

Mitochondrial rewiring supports survival of triple negative breast cancer cells after ionizing radiation

Triple negative breast cancer (TNBC) is an aggressive disease with limited therapeutic options. Conventional treatments include neoadjuvant chemo-immunotherapy followed by surgical resection and may include further adjuvant immunotherapy and/or radiotherapy of the tumor bed and lymph nodes. Nonetheless, TNBC patients with residual disease have rapid metastatic recurrence. While the roles of metabolic and mitochondrial adaptations in chemotherapeutic resistance have been the subject of many studies, their importance in the context of ionizing radiation (IR) therapy remains poorly understood. We established longitudinal in vitro models of post-IR human TNBC, characterized by cellular regression to a residual phenotypic state, then eventual cell repopulation. This was accompanied by plastic adoption of unique metabolic, proteomic, and morphologic features that largely reverted when cells regrew. Following IR, residual cells exhibited extensive mitochondrial rewiring, including elevated mitochondrial content, oxidative phosphorylation (oxphos) rates, cristae structures, and metabolite levels. Concomitantly, levels of the short protein isoform of the mitochondrial inner membrane protein optic atrophy 1 (OPA1) were significantly elevated in residual cells, and OPA1 knockout ablated mitochondrial adaptations induced by IR. OPA1 genetic or pharmacologic perturbations led to improved cellular responses to IR. Metabolomic and proteomic analyses of radio-residual cells uncovered a coordinated program of antioxidant and redox capacity elevation with mitochondrial metabolism, which was corroborated by analyses of external datasets. Together, these findings provide evidence that TNBC cells surviving radiotherapy adopt an OPA1-dependent program of mitochondrial reorganization that supports their survival and regrowth, thereby positioning OPA1 as a therapeutic dependency that could improve radiotherapy efficacy in TNBC.

cancer biology↗

bioRxiv2026

Toxoplasma GRA8 engages the host ESCRT accessory protein ALG-2 and is necessary for parasite metabolic integrity

The endosomal sorting complex required for transport (ESCRT) is a hetero-multimeric membrane-remodeling machinery essential for endosomal sorting, intraluminal vesicle formation, cytokinetic abscission, and membrane repair. ESCRT is also hijacked by some pathogens including the protozoan Toxoplasma gondii, which subverts it at the parasitophorous vacuole membrane to support parasite ingestion of host cytosolic proteins. Although the ESCRT accessory protein, ALG-2 is recruited to the parasitophorous vacuole, nothing was known about how this happens. Herein we identify the dense granule protein TgGRA8 as a key effector that recruits ALG-2 and ALIX to the parasitophorous vacuole. We show that TgGRA8 directly binds ALG-2 via conserved ALG-2-binding elements like those found in other ALG-2 interacting proteins including ALIX and SEC31A. Biochemical assays show high-affinity, Ca{superscript 2}-dependent TgGRA8-ALG-2 binding, and structural modeling suggests TgGRA8 may assemble multivalently to coordinate multiple ALG-2 dimers, stabilizing ALG-2/ALIX recruitment through a non-canonical bridging mechanism. Conservation of these motifs among tissue cyst-forming coccidians implies a lineage-linked adaptation that supports infection by these parasites. Metabolomics further indicates that TgGRA8 loss disrupts amino acid, purine, and central carbon metabolism, like those seen in other ingestion deficient mutants. Together, these findings uncover a conserved, multivalent strategy by which Toxoplasma engages host ALG-2 to organize ESCRT at the parasitophorous vacuole, thereby coupling nutrient acquisition to metabolic fitness and exposing a novel agent for probing ESCRT biology.

cell biology↗

bioRxiv2026

FOXO/DAF-16 modulates the transcription factor ROR/NHR-23 and inhibits the let-7 microRNA to maintain multipotency during dauer

Animals rapidly reprogram gene expression to adapt development to environmental stress. How gene regulatory programs that drive continuous development are repressed during stress-induced developmental arrest remains poorly understood. In Caenorhabditis elegans, starvation and overcrowding trigger entry into the stress-resistant, quiescent dauer stage. Here, we identify interactions among the conserved transcription factors DAF-16/FOXO and NHR-23/ROR, and the let-7 family of microRNAs as key regulators of the switch from continuous development to dauer. We show that loss of daf-16 during dauer causes elevated let-7 family microRNAs and premature expression of the adult collagen reporter col-19p::GFP. Reducing let-7 family activity suppresses this phenotype, whereas dauer-specific let-7 expression is sufficient to induce col-19p::GFP expression. Mechanistically, DAF-16 inhibits let-7 transcription in part by repressing nhr-23, which encodes a transcriptional activator of the let-7 family and molting-cycle genes. ChIP-seq analysis reveals DAF-16 binding upstream of nhr-23, and daf-16; daf-7 mutant dauers exhibit increased nhr-23 mRNA and NHR-23 protein, supporting a model in which DAF-16 directly represses nhr-23. Integrated ChIP-seq and transcriptomic analyses identify 1,183 genes activated and 681 genes repressed by DAF-16 during dauer. Repressed targets are enriched for pro-growth genes involved in mitotic DNA replication and translational elongation. DAF-16 targets include 59 transcription factors that may mediate broader transcriptional reprogramming during dauer to maintain multipotency and establish quiescence. Together, these findings reveal that DAF-16/FOXO establishes stress-induced developmental arrest by coupling activation of protective pathways with repression of conserved developmental timing, growth, and differentiation programs. Significance statementAnimals often pause development during environmental stress and then resume normal development when conditions improve. How developmental programs are temporarily halted without disrupting later cell fates remains poorly understood. We show that, during stress-induced dauer arrest in C. elegans, the conserved FOXO transcription factor DAF-16 represses the differentiation-promoting let-7 microRNA pathway. DAF-16 inhibits the ROR homolog NHR-23, which normally activates let-7 and molting-cycle genes. Genome-wide analyses further show that DAF-16 both activates stress-response genes and represses growth and developmental genes. These findings reveal how a conserved stress-responsive factor coordinates survival, developmental arrest, and maintenance of cellular multipotency.

molecular biology↗

bioRxiv2026

Substrate biasing in UCHL5 proteoforms

Proteolytic deubiquitinating enzymes bridge a gap in substrate recognition through complex regulatory mechanisms. A growing portion of these are accomplished through proteoforms that uniquely control association and diverse sets of cleavage capabilities that relay distinct physiological outcomes. This study describes substrate biasing governed by UCHL5 proteoforms. It demonstrates that N-terminal ubiquitination activates the enzyme towards monoubiquitin substrates, a feature that is conserved across UCHL5 homologs. Crystallographic and spectroscopic data suggest that the N-terminal ubiquitin binds intramolecularly in an allosteric binding site and inhibits branched chain substrate cleavage. Association with Rpn13/Adrm1 relieves this inhibition and reestablishes its ability to debranch, potentially controlling nonspecific debranching compared to retention of needed activity on the 26S proteasome. Collectively, this study describes the molecular basis for substrate selectivity in a deubiquitinating enzyme, an unexplored area in the enzymes that counteract ubiquitin E3 ligases.

biochemistry↗

bioRxiv2026

Recommendations for the ethical and accurate use of population descriptors: a trainee-led survey of early-career researchers

Despite the importance of population descriptors in human genomics research, many scientists struggle to translate evolving ethical guidelines into their computational workflows. To characterize this gap between recommendations and implementation, we conducted a mixed-methods survey of early-career researchers to assess how they understand and implement the landmark 2023 NASEM report on the use of population descriptors in human genetics research. We show that while exposure to the report fosters ethical awareness, fundamental misconceptions about race and ancestry persist across academic disciplines, and trainees face structural bottlenecks, including legacy data constraints and a lack of technical confidence. To address this gap, we offer actionable, stakeholder-specific recommendations across the research lifecycle ranging from decision-support tools to "bring-your-own-data" workshops to leadership from academic journals, scientific societies, and trainee mentors. Ultimately, we argue that to promote scientific rigor and reduce bias in genetic discoveries, the scientific ecosystem must invest in the infrastructure necessary to empower the next generation of researchers.

genetics↗

bioRxiv2026

DNA Sequence and Histone Variant H2A.Z Jointly Govern Nucleosome Unwrapping Pathways

Nucleosome unwrapping governs chromatin accessibility and gene regulation, yet the molecular determinants of unwrapping directionality remain poorly understood. Using atomistic and SIRAH coarse-grained umbrella sampling simulations, we show that DNA sequence and histone variant composition jointly tune a directional preference for nucleosome unwrapping. For both the ASP and Widom-601 sequences, unwrapping initiates asymmetrically from a preferred DNA end, with progressive disengagement of the H3 N-terminal tail providing the molecular switch that determines directionality in the Widom-601 system. Substitution of canonical H2A with the variant H2A.Z reverses this directional preference, shifting unwrapping to the opposite DNA end and altering the free energy landscape. SIRAH coarse-grained simulations faithfully reproduce these sequence- and variant-dependent unwrapping pathways and their qualitative free energy features, though quantitative barrier heights differ from atomistic values, identifying a target for further force field refinement. Comparing the H3 tail flexibility from atomistic simulations with published solution NMR amide intensities for two Widom-601 constructs corroborates the fast-timescale tail dynamics and points to a sequence-dependent, microsecond-millisecond exchange component at the H3 tail-core junction. Together, these results establish H3 tail - DNA disengagement, corroborated by NMR data, as a key mechanistic determinant of unwrapping directionality, reveal how a single histone variant substitution can reverse this preference, and validate SIRAH as an efficient framework for large-scale chromatin simulations.

biophysics↗

bioRxiv2026

Identification, optimization, and structural elucidation of chloroacetamide scaffold as covalent inhibitors for Ubiquitin C-terminal Hydrolase L3

The deubiquitinating enzyme, ubiquitin C-terminal hydrolase L3 (UCHL3), has been implicated as a potential therapeutic target for cancer with a role in regulating the DNA damage response pathways. While the target has been studied using genetic methods there is a lack of reliable chemical probes to selectivity target UCHL3. In this study we report hit identification and optimization of a new chemical scaffold that irreversibly inhibits UCHL3. The observed structure-activity relationships are corroborated by ligand-bound crystal structures that confirm covalent adduct formation with the catalytic cysteine of the enzyme. Finally, through gel-shift assays using a ubiquitin activity-based probe we demonstrate on-target engagement with UCHL3 in two cell lines. The work as a whole presents a comprehensive evaluation of the new scaffold that can be utilized to probe UCHL3 in different biological contexts.

pharmacology and toxicology↗

bioRxiv2026

Phylogenetic Analysis and Structural Evaluation of Staphylococcus aureus Serine-Aspartate Repeat-Containing Protein D with a Focus on Periprosthetic Joint Infection

Serine-aspartate repeat-containing protein D (SdrD) is a Staphylococcus aureus cell wall-anchored, calcium-binding adhesin member of the MSCRAMM Sdr subfamily that may contribute to bacterial adhesion and virulence. S. aureus is the most common cause of periprosthetic joint infection (PJI). Population-level distribution and sequence diversity of SdrD among clinical PJI isolates have not been systematically characterized, and the SdrD binding mechanism is still not well understood. To address these gaps, sdrD alleles were queried across 156 newly sequenced PJI isolates and compared to publicly available S. aureus genomes, and nucleotide- and protein-level phylogenies of the sdrCDE locus constructed. The SdrD crystal structure from S. aureus JH1 was determined, with solution small-angle X-ray scattering (SAXS) and molecular dynamics (MD) simulations, and assessment of conformational changes with calcium depletion. Three dominant sdrD subtypes were defined, associating with USA300, JH1, and TCH60; the JH1 sdrD subtype was predominant among PJI isolates. Structural studies showed that the conformation of individual domains and interdomain organization of the multidomain SdrD have limited flexibility in solution, and that the calcium-binding B domain retains its core fold under conditions of calcium depletion. Together, the findings presented support functional diversification among Sdr family members in mediating host attachment and inform a re-evaluation of the ligand-binding mechanism previously proposed for SdrD. AUTHOR SUMMARYStaphylococcus aureus is the leading cause of infections that develop around joint implants (periprosthetic joint infection, PJI). This bacterium has a large arsenal of surface proteins that allow it to stick to human tissues and implanted devices. This work focused on one such protein, SdrD, which has been linked to implant-associated infections but the structure and diversity of which among patients with PJI had not been well characterized. The genetic sequences of SdrD were analyzed across thousands of bacterial genomes, including those from patients with PJI. Distinct genetic variants of the protein were found, one of which was particularly common with PJI. The three-dimensional structure of SdrD was determined at atomic resolution and solution small-angle X-ray scattering (SAXS) and molecular dynamics used to study how it moves and responds to changes in its environment. Contrary to what was previously described, SdrD was shown to be relatively rigid. These findings change how SdrDs mechanism of action should be considered, potentially informing design strategies to block bacterial attachment before infection takes hold.

biophysics↗

bioRxiv2026

Establishment of snake venom gland organoids from a novel family, Colubridae

Non-front-fanged snakes are abundant, diverse and represent approximately 70% of extant snakes. However, there is limited knowledge about most species and their venoms, in part due to the technical and welfare challenges associated with venom extraction, low venom yields, and the lack of cellular models available. Organoids represent an excellent opportunity to overcome these challenges. Here, we establish, for the first time, venom gland organoids from snakes of the Colubridae family and demonstrate the in vitro production of toxins.

bioengineering↗

bioRxiv2026

AlphaFold Database expands to proteome-scale quaternary structures

Protein function is governed by molecular interactions, yet structural coverage of these interactions remains sparse. The AlphaFold Protein Structure Database (AFDB) transformed access to accurate monomeric protein structures at scale. Here, we expand the AFDB to quaternary structures by predicting 31M candidate homo- and heterodimeric protein complexes, compiled from 4,777 proteomes, including model- and global health organisms. We established confidence criteria via analysis of experimentally determined structures, resulting in 1.81M high-confidence predictions. These models enabled the discovery of emergent structures and topologies not present in monomeric predictions. Additionally, the top 1% of structural clusters accounted for [~]44% of all complexes, and [~]8.3% of clusters were conserved across multiple domains of life, pointing to a substantial fraction of ancient, universally retained assemblies. Structural search highlighted that high-confidence predictions are anchored in experimental multimer space, yet 31.3% extend beyond detectable PDB coverage. These freely accessible proteome-scale predictions facilitate functional and mechanistic hypothesis generation across biology.

synthetic biology↗

bioRxiv2026

Endothelial ACKR1 expression regulates neutrophil infiltration and breast cancer metastatic engraftment in the lung metastatic niche

The formation of the premetastatic niche prepares distant tissues for tumor cell engraftment. Endothelial cells are critical mediators of premetastatic niche formation, orchestrating extravasation of circulating tumor cells and critical pro-tumor immune cells, such as neutrophils. In mouse models of breast cancer, we show that primary tumors upregulate the non-signaling chemokine receptor ACKR1 in the endothelium of the lung premetastatic niche. ACKR1-expressing venules were found to be preferential sites of neutrophil and tumor cell localization within lung tissue. A newly generated conditional ACKR1 allele was used to show that endothelial-specific removal of ACKR1 expression significantly reduces metastatic engraftment in the lung. When ACKR1 is activated by tumor-secreted factors, endothelial ACKR1 functions to promote neutrophil recruitment within the lung parenchyma. We conclude that ACKR1 is a critical component of the endothelial response to tumors at the metastatic site of the lung, leading to neutrophil recruitment and promotion of tumor cell metastasis. SUMMARYEndothelial cells play critical roles in breast cancer metastasis. ACKR1 is upregulated in the endothelium of the lung metastatic niche in response to primary mammary tumors. Endothelial ACKR1 expression was found to promote neutrophil infiltration into the metastatic niche and support breast tumor cell metastasis to the lung.

cancer biology↗

bioRxiv2026

Oxidative Stress-Induced Immunogenic Cell Death Enhances Whole-Cell Vaccine Efficacy in a Syngeneic Pancreatic Cancer Model

Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive cancer, with limited therapeutic options and extremely high mortality rates. While immune checkpoint blockade (ICB) therapy is effective in many types of human cancers, responses in PDAC patients remain poor, partly due to the weak immunogenicity of PDAC tumors. We hypothesized that a whole-cell PDAC vaccine could improve anti-tumor responses if optimized to expose a more stimulatory repertoire of tumor antigens. To test this, we used murine Panc02 pancreatic cancer cells to screen several stress-inducing treatments (UV, hypoxia, heat shock, and hydrogen peroxide [H2O2]), among which low-dose oxidative stress (0.05% H2O2 for 2h) was identified as the optimal inducer of immunogenic cell death (including increased surface calreticulin, ERp57 exposure, HMGB1 release and MHC class I expression). We then prepared a whole-cell vaccine of fixed H2O2-treated Panc02 cells, which induced robust tumor-specific immunity in C57BL/6 mice bearing syngeneic Panc02 tumors. Vaccine-treated mice displayed a significant increase in tumor-reactive IFN{gamma}+ T cells, as well as extensive tumor infiltration by CD4 + and CD8 + T cells and NCR1+ NK cells. When used prophylactically, the vaccine significantly delayed tumor growth and extended survival, whereas therapeutic application markedly slowed tumor progression. Importantly, combining the whole-cell Panc02 vaccine with anti-PD-1 therapy induced complete tumor regression in a subset of animals. Together, these data demonstrate that controlled oxidative stress can convert autologous tumor cells into an effective whole-cell vaccine without the need for genetic modification or prior neoantigen identification, offering a scalable strategy for personalized immunotherapy in PDAC. STATEMENT OF SIGNIFICANCEThis study demonstrated that oxidative stress-induced immunogenic cell death reprograms pancreatic tumor cells to induce danger signaling and enhance antigen presentation, thereby promoting immune infiltration and sensitizing tumors to PD-1 blockade.

cancer biology↗

bioRxiv2026

AI System Using Unsupervised Learning to Discover Novel Subtypes in Alzheimer's Disease

Early Alzheimers disease often evades timely detection because typical diagnostics are based on symptomatic thinking rather than intrinsic neurodegeneration. Here, we use unsupervised machine learning to identify latent Alzheimers phenotypes from structural MRI-derived volumetric features and neuropsychological scores, without using diagnosis labels or predefined subtype definitions. We analyzed participants (18-96 years) from the OASIS-1 study using intracranial-normalized global and regional volumetric MRI features together with MMSE and CDR measurements. After dimensionality reduction with principal component analysis, we identified five stable clusters using K-Means clustering. Here, one cluster exhibited salient cortical atrophy but intact preserved cognitive function, indicative of an independent preclinical subtype. As a reproducibility check, random forest and logistic regression models trained to predict cluster membership achieved >90% cross-validated accuracy, indicating that the clusters were consistently separable in the learned feature space. Age and education did not fully explain this structural-functional dissociation, suggesting a subgroup with relative cognitive resilience despite measurable atrophy. Our findings challenge the assumption of a uniform atrophy-cognition relationship and suggest that data-driven phenotyping may reveal clinically relevant subgroups not captured by conventional diagnostic frameworks. Future work will apply validation to longitudinal cohorts, in addition to incorporating multimodal biomarkers.

neuroscience↗