Biology subjects

Grant, E. T.

Publications and source records attributed to Grant, E. T..

5 recordsLinked to original sources

bioRxiv2026

Language models annotate mass cytometry clusters within the range where human experts disagree

Cell-type identification of clusters from unsupervised cytometry pipelines (e.g. FlowSOM, GigaSOM) is a time-consuming, expert-dependent bottleneck, typically requiring an expert to read a heatmap of median marker intensities and assign a label to each cluster. Large language models (LLMs) can annotate single-cell RNA-seq clusters from marker-gene lists, but cytometry cluster data differ in both dimensionality and data type. We present a direct procedure in which cluster medians are scaled per marker and grouped by similarity into a few dozen metaclusters. The continuous marker profiles are then passed to the LLM, which returns a cell-type label and confidence level for each metacluster in a single call, without a gating hierarchy, manually defined marker-specific thresholds, or task-specific training data. Across four human and mouse datasets spanning blood, bone marrow and gut, a single well-chosen model reproduces the reference annotation at the major-lineage level with 85-98% metacluster-level concordance (cloud Claude Sonnet 4.6 85-98%, best local model 83-96%), substantially exceeding majority-class baselines; chance-corrected agreement is Cohen's k = 0.67-0.83. Local open-weight models perform competitively with the cloud model on the independent benchmarks, enabling application to clinical or patient-derived data subject to data-protection restrictions. Reliability is model-dependent: no local model is best everywhere, and the frontier model Claude Opus 5 was no more reliable than Claude Sonnet 4.6, the most consistent in concordance across datasets. Because no annotation is absolute ground truth, we obtained two additional expert annotations of the same continuous heatmaps, blinded to the reference and to all model output, to establish a method-matched agreement ceiling. The experts agreed on ~99% of major-lineage assignments but only ~73% of T-cell maturation states; on that axis the best models fall within the expert disagreement range. LLM-based annotation is therefore a practical, training-free alternative to rule-based annotation, provided its output is checked against the intended panel and species.

bioinformatics↗

bioRxiv2025

A prebiotic-postbiotic combination supports dietary carbohydrate-targeting functional properties in a fiber-deprived microbiota

Dietary fiber deprivation compromises gut mucosal barrier integrity by promoting microbial degradation of host mucus, a process linked to various gut-related auto immune diseases. While postbiotics are considered safer alternatives to fiber for susceptible patients, their mechanistic ef- fects on a fiber-deprived gut remain poorly understood. Here, we demonstrate in a mouse model that a fermented postbiotic, alone or in combination with a prebiotic and aloe vera, counteracted the increase of detrimental properties of the microbiota on a fiber-free diet. The supplement regimen reshaped the gut microbiota, counteracting the expansion of key mucin-degrading bac- teria, including Akkermansia muciniphila and Parabacteroides goldsteinii. Metatranscriptomic analysis revealed this compositional change corresponded to a community-wide functional pivot away from expressing mucinolytic enzymes, such as sialidases, and towards utilizing alternative substrates. These microbial shifts recapitulated the effects of dietary fiber reintroduction and translated to direct host benefits, including sustentation of the colonic mucus layer and attenu- ation of diet-induced type III immune cytokine expression. Our findings provide a mechanistic rationale for using postbiotics to functionally replace dietary fiber, offering a promising strategy to support gut homeostasis in contexts where fiber intake is limited.

microbiology↗

bioRxiv2023

Maternal diet and gut microbiome composition modulate early life immune responses

In early life, the intestinal mucosa and immune system undergo a critical developmental process to contain the expanding gut microbiome while promoting tolerance towards commensals, yet the influence of maternal diet and gut microbial composition on offspring immune maturation remains poorly understood. We colonized gnotobiotic mice with a defined consortium of 14 strains, fed them a standard fiber-rich chow or a fiber-free diet, and then longitudinally assessed offspring development during the weaning period. Unlike pups born to dams fed the fiber-rich diet, pups of fiber-deprived dams demonstrated delayed colonization with Akkermansia muciniphila, a mucin-foraging bacterium that can also utilize milk oligosaccharides. The pups of fiber-deprived dams exhibited an enrichment of colonic tissue transcripts corresponding to defense response pathways and a peak in Il22 expression at weaning. Removal of A. muciniphila from the community, but maintenance on the fiber-rich diet, was associated with reduced proportions of ROR{gamma}t-positive innate and adaptive immune cell subsets. Our results highlight the potent influence of maternal dietary fiber intake and discrete changes in microbial composition on the postnatal microbiome assemblage and early immune development.

microbiology↗

bioRxiv2022

Increased gut microbial mucin foraging promotes clearance of a parasitic worm

BACKGROUND & AIMSHost-secreted gastrointestinal mucus plays a key role in the expulsion of intestinal nematode parasites. A balance between mucin secretion by the host and the gut microbial mucin foraging is essential to maintain the intestinal homeostasis, yet little is known about how changes in the mucin-microbiome interactions affect worm infections. Here, we aimed to examine how mucin foraging activity by the microbiome changes the course of parasitic worm infections by modulating the host immune responses. METHODSWe utilized a gnotobiotic mouse model containing a synthetic human gut microbiota that allows for: 1) a complete removal of the mucin-degrading bacteria from the community; and 2) diet-driven manipulation of the microbiota toward mucin foraging. We infected mice with a murine nematode, Trichuris muris, which resembles human infection with Trichuris trichiura. We examined the temporal dynamics of worm infection including worm burden and the host immune responses, and coupled these readouts to the microbial changes and mucin foraging activity. RESULTSThe absence of mucin-degrading bacteria in the microbiota enhances susceptibility to parasitic infection--evidenced by higher worm number--by promoting stronger Th1 immune responses. Dietary fiber deprivation increases the microbial mucin-foraging activity, which coincides with a shift in host immune responses from susceptible (chronic, Th1 type) to resistant (acute, Th2 type), thereby promoting worm clearance. CONCLUSIONSOur results provide mechanistic insights into how the colonic mucin-degrading bacteria promote anti-parasitic immunity through modulation of the host immune responses. Our study documents a clinically-relevant, novel link in the microbiome-parasite-host immune axis that is useful prerequisite knowledge in managing parasitic infections.

immunology↗

bioRxiv2021

Deprivation of dietary fiber in specific-pathogen-free mice promotes susceptibility to the intestinal mucosal pathogen Citrobacter rodentium

The change of dietary habits in Western societies, including reduced consumption of fiber, is linked to alterations in gut microbial ecology. Nevertheless, mechanistic connections between diet-induced microbiota changes that affect colonization resistance and enteric pathogen susceptibility are still emerging. We sought to investigate how a diet devoid of soluble plant fibers impacts the structure and function of a conventional gut microbiota in specific-pathogen-free (SPF) mice and how such changes alter susceptibility to a rodent enteric pathogen. We show that absence of dietary fiber intake leads to shifts in the abundances of specific taxa, microbiome-mediated erosion of the colonic mucus barrier, a reduction of intestinal barrier-promoting short-chain fatty acids, and increases in markers of mucosal barrier integrity disruption. Importantly, our results highlight that these low fiber diet-induced changes in the gut microbial ecology collectively contribute to a lethal colitis by the mucosal pathogen Citrobacter rodentium, which is used as a mouse model for enteropathogenic and enterohemorrhagic Escherichia coli (EPEC and EHEC, respectively). Our study indicates that modern, low-fiber Western diets might make individuals more prone to infection by enteric pathogens via the disruption of mucosal barrier integrity by diet-driven changes in the gut microbiota, illustrating possible implications for EPEC and EHEC infections.

microbiology↗