Biology subjects

Le, M.

Publications and source records attributed to Le, M..

15 recordsLinked to original sources

bioRxiv2026

Interpretable spherical geometry of single-cell state transitions from dominant principal components

Single-cell RNA-seq atlases are commonly explored with nonlinear embeddings that preserve neighborhoods but provide limited coordinate-level interpretation. We asked whether projecting the dominant principal components (PCs) of single-cell gene expression onto a unit sphere would yield an interpretable coordinate system. SPHERE-PCA L2-normalizes the first three PC coordinates, aligns a biologically defined root to the north pole, and represents each cell by three coordinates: root-aligned geodesic distance ({theta}), angular position ({phi}), and pre-projection radial magnitude (r). Across developmental and disease-associated datasets, this representation reveals structured spherical geometry, ranging from near-great-circle trajectories to multi-arc manifolds. In developmental atlases, root-aligned geodesic distance increases as CytoTRACE-inferred stemness decreases, while gene-coordinate analyses separate programs associated with angular position from those associated with radial magnitude. Fixed-loading perturbations decompose each gene's effect on cell position into progression, branch- or state-position, and radial activity components. SPHERE-PCA therefore provides a deterministic, loading-preserving coordinate framework for interpreting dominant transcriptomic variance and establishing a transparent geometric coordinate framework for perturbation analysis and virtual-cell models.

systems biology↗

bioRxiv2026

Reward Memories Bias Instrumental Rule Selection via the Orbitofrontal-to-Secondary Motor cortex Pathway

Reward-predictive cues can influence decision-making and promote the pursuit of specific outcomes. This influence is classically studied using specific Pavlovian-to-instrumental transfer (sPIT), in which Pavlovian cues bias action selection and promote the instrumental action (often a left or right lever press) directed toward the cued outcome. However, naturalistic reward pursuit often extends beyond selecting discrete actions and requires selecting abstract rules that organize multiple actions into goal-directed sequences. The influence of cued reward memories on such rule selection has received less attention, and the neural circuits mediating this effect remains largely unknown. Here, we developed the specific Pavlovian-to-Rules-to-Instrumental Transfer (sPRInT) task, an adaptation of sPIT designed to examine how Pavlovian reward cues bias instrumental rule selection. We then used pathway-specific chemogenetic silencing to test the contribution of orbitofrontal cortex projections to secondary motor cortex (OFC[->]M2) to this effect. Rats expressing hM4Di or mCherry in OFC[->]M2 neurons learned a three-step instrumental sequence (sample lever [->] nosepoke [->] choice lever). In alternating blocks, rats used either a delayed non-match-to-sample rule or a visually guided rule to select the final action in the sequence and earn distinct outcomes (dNMTS-O1; VIS-O2). In a second phase, two distinct auditory cues were established as Pavlovian predictors of the two outcomes (S1-O1; S2-O2). Finally, in nonrewarded probe tests preceded by DCZ injections, we examined how presentation of these Pavlovian cues biased instrumental choices by promoting the use of an abstract rule. In control (mCherry) rats, Pavlovian cues promoted the adoption of the rule corresponding to the cued outcome (S1 promoted dNMTS; S2 promoted the VIS rule). In hM4Di rats, DCZ-mediated inactivation of OFC[->]M2 projection neurons diminished this effect. In contrast, OFC[->]M2 inhibition spared conventional sPIT, in which reward-predictive cues directly biased action selection without requiring abstract rules or extended action sequences. These findings demonstrate that cued reward memories can govern abstract rule selection and identify OFC[->]M2 as a critical circuit for translating those memories into goal-appropriate behavioral strategies.

neuroscience↗

bioRxiv2026

Sharp and Fast Dynamic Extraction and Tracking of Emitted Cellular Transients

Genetically encoded fluorescent sensors have expanded our ability to image cellular activity and transmitter release. Yet, sparse and low-salience events remain difficult to resolve against complex and fluctuating fluorescence backgrounds. Here we introduce DETECT, Dynamic Extraction and Tracking of Emitted Cellular Transients, which combines adaptive background suppression, probabilistic classification and multi-object tracking to extract fluorescence events while preserving their identity. Across synthetic datasets, DETECT improved detection and segmentation accuracy and reduced computational cost relative to established event-based methods. We validated DETECT across confocal, two-photon and miniscope imaging, ex vivo and in vivo, using calcium indicators and monoamine sensors. Beyond its technical performance, DETECT extends event-based analysis to low-salience fluorescence signals while resolving events spanning broad ranges of amplitude, morphology and dynamics. By resolving spontaneous dopamine and noradrenaline signals as distinct, trackable release events, DETECT reveals the spatiotemporal organization of neuromodulatory activity and provides a broadly applicable approach to quantitative fluorescence analysis.

neuroscience↗

bioRxiv2026

Orbitofrontal circuits for context-gated reward predictions

The orbitofrontal cortex (OFC) is essential for the contextual regulation of reward predictions, yet the circuit mechanisms underlying this function remain unclear. Here, we dissected the roles of two major OFC efferent pathways --projections to the central dorsal striatum (CDS) and the mediodorsal thalamus (MDT)-- in contextual regulation. Male and female rats performed a context-dependent discrimination task in which the reward-predictive status of two auditory target cues (X and Y) was determined by a visual contextual cue (A). Using temporally precise optogenetic inhibition of OFC terminals, we found that silencing OFC[->]CDS profoundly disrupted contextual gating, virtually abolishing the negatively gated component (A:Y- / Y+) while producing modest effects on the positively gated component (A:X+ / X-). In contrast, OFC[->]MDT inhibition produced modest and qualitatively distinct effects, characterized by a generalized increase in responding in context A and a decrease in context noA. These effects were reproduced in a connectionist model in which context influenced predictions either additively (via elemental representations) or by acting as a gatekeeper (via configural cue/context representations). Disrupting the configural pathway mimicked OFC[->]CDS silencing, whereas amplifying additive contextual inputs replicated OFC[->]MDT inhibition. Together, these findings reveal complementary circuit mechanisms through which the OFC shapes context-appropriate reward seeking.

neuroscience↗

bioRxiv2026

Emergence of mpox in Guinea: sporadic clade IIa cases and a clade IIb upsurge

BackgroundMpox virus (MPXV) has caused recurrent outbreaks in West Africa. However, Guinea had not previously reported laboratory-confirmed cases or MPXV genomic data. MethodsSuspected cases were identified in the NZerekore region as well as in the Conakry (Sept 2024- Dec 2025) and confirmed by real-time PCR in regional and central laboratories, respectively. Whole-genome sequencing using nanopore technology was performed in-country, followed by phylogenetic and time-scaled evolutionary analyses. FindingsThe first mpox case was clinically diagnosed in September 2024 in the NZerekore region and laboratory-confirmed by the prefectural laboratory in Gueckedou. In total, seven cases were confirmed in Forest Guinea, of which five complete or almost complete MPXV genomes were recovered. All belonged to MPXV clade IIa. Genomic divergence, ancestral dating, and low APOBEC3-associated mutation frequencies were consistent with multiple independent zoonotic spillover events. In June 2025, one of the first mpox cases of an unfolding outbreak was confirmed in Conakry. Whole genome sequencing revealed MPXV clade IIb lineage G.1. By December 2025, the number of laboratory-confirmed mpox cases nationwide increased to 2,151. A total of nine outbreak strains were sequenced, all belonging to Clade IIb. The genomes clustered with contemporaneous genomes from Sierra Leone and showed high APOBEC3-associated mutation frequencies, suggesting sustained human-to-human transmission in the region. InterpretationThese data demonstrate simultaneous circulation of MPXV clade IIa and IIb strains in Guinea, likely resulting from zoonotic spillover and human-to-human transmission, respectively. Decentralised diagnostics and in-country sequencing facilitated rapid case confirmation and genomic surveillance, highlighting the importance of these critical capacities for outbreak preparedness and response. FundingGerman Federal Ministry of Health and the German Center for Infection Research (DZIF).

genomics↗

bioRxiv2026

PARP1 recruits SPRTN to DNA-protein crosslinks through a conserved poly-ADP-ribose binding domain.

DNA-protein crosslinks (DPCs) are toxic DNA lesions formed by the covalent attachment of proteins to DNA. Failure to resolve DPCs leads to genomic instability, premature aging, and cancer predisposition. Although multiple proteases and the 26S proteasome degrade DPCs, how these lesions are detected and marked for proteolysis remains unclear. Here, we show that poly-(ADP-ribose) polymerases (PARP1/2) sense DPCs and modify them with poly(ADP-ribose) (PAR) to promote repair via a SPRTN-Tdp1 axis. We discovered a Nudix homology domain (NHD) in SPRTN that mediates direct non-covalent PAR binding and is important for DPC repair. Loss of PARP1/2 activity or mutation of the SPRTN NHD leads to sustained DPCs. Single-molecule analysis revealed that SPRTN does not bind efficiently to the DPC, however after the addition of PARP1 in the presence of NAD+, SPRTN binding to the DPC was significantly increased. Our findings establish PARP1/2 enzymes as immediate DPC sensors, reveal PARylation as a signal marking DPCs for SPRTN-dependent degradation, and identify SPRTN as the first PARP-directed protease.

molecular biology↗

bioRxiv2025

Proteostasis Stress Drives Stem Cell Aging, Clonal Hematopoiesis and Leukemia

Aging is the primary risk factor for clonal hematopoiesis and the development of hematologic malignancies (1-5), yet the selective pressures that shape stem cell behavior and clonal expansion during aging remain poorly defined. Here, we identify proteostasis stress as a central driver of hematopoietic stem cell (HSC) aging and clonal evolution. We show that Heat shock factor 1 (Hsf1) is activated in aging HSCs to preserve proteostasis and sustain self-renewal. However, this physiological, age-associated adaptive mechanism is co-opted by pre-leukemic Dnmt3a-mutant HSCs to resist proteostasis and inflammatory stress required to fuel clonal expansion during aging. In the context of co-occurring Dnmt3a and Nras mutations, which are frequently observed in human acute myeloid leukemia (AML) (6-13), mutant HSCs and progenitors exhibit heightened dependence on Hsf1 for expansion, malignant transformation and disease progression. Loss of Hsf1, or disruption of proteostasis, impairs expansion of mutant progenitors, delays leukemia onset, and prolongs survival. Together, these findings reveal proteostasis as a key constraint in the aging hematopoietic system that imposes a selective bottleneck. Hsf1 activation enables both physiological adaptation in aging stem cells and pathological clonal outgrowth in pre-leukemic and leukemic states, establishing proteostasis control as a pivotal mechanism linking stem cell aging to clonal hematopoiesis and malignancy.

cancer biology↗

bioRxiv2025

Reproducibility and Accuracy of Nanopore-Based Methylome Profiling of Streptococcus dysgalactiae subspecies equisimilis Strains from Cancer Patients

BackgroundDNA methylation influences bacterial gene regulation, virulence, and restriction-modification (RM) systems. Advances by Oxford Nanopore Technologies (ONT) now enable direct methylome profiling from nanopore sequencing using the Dorado basecaller. However, the comparative performance of ONT-only versus hybrid-assembly reference-based methylation calling, particularly regarding genomic DNA quality and inter-operator variability, remains understudied. MethodsSix operators independently prepared fifteen sequencing libraries each for nanopore (MinION R10.4.1 flow cells, Mk1D) and Illumina MiniSeq platforms for two Streptococcus dysgalactiae subsp. equisimilis strains (UT9728, 12 replicates; UT10237, 3 replicates). MicrobeMod v1.0.3 was used to identify methylation and motif profiles using Illumina-corrected hybrid reference assemblies (HRAs) and ONT-only reference assemblies (ORAs). Reproducibility and accuracy were compared using a custom genome annotation feature-enabled modular analysis that mapped and counted methylation site calls to CDS, rRNA and tRNA coordinates. ResultsStrain UT9728 predominantly exhibited N6-methyladenine (6mA) at GATC motifs, whereas strain UT10237 displayed dual methylation patterns: C5-methylcytosine (5mC) at CCWGG motifs and 6mA at GAGNNNNNTAA motifs. Both strains contained Type I and Type II RM systems; UT10237 uniquely harbored a Type IIG RM system with combined restriction and methylation activities. Motif identification concordance using HRAs and ORAs exceeded 99.9%. Reproducibility for methylation calls was high across independent replicates for both HRA (Pearsons r >0.989) and ORA (Pearsons r >0.993) methylation calls in GATC and CCWGG motifs but lower in the GAGNNNNNTAA motif (Pearsons r (HRA) = 0.80; r (ORA) = 0.78). ORA-based methylation site calls for all motifs showed excellent precision and recall compared to HRA-based calls (F1-score >99.999%). ConclusionOur findings support the accuracy, robustness, and utility of ONT-only data based methylome profiling for bacterial epigenetic characterization. Our analytical framework facilitates detailed evaluations of reproducibility and accuracy. Data Summary and AvailabilityO_LIUnprocessed_data_files: DOI: 10.5281/zenodo.15555625 O_LIRaw paired-end FastQ Files (Illumina). C_LIO_LIRaw FastQ Files (ONT; unmodified basecalls). C_LIO_LIRaw uBAM Files (ONT; 6mA_5MC_modified_basecalls). C_LI C_LIO_LIGenome_assemblies_and_annotation_files: DOI: 10.5281/zenodo.15558488 O_LIGenome_Assemblies reconstucted with ONT reads and polished with Illumina Reads (Hybrid assemblies);(strain_name_ _Hyb.gbk) C_LIO_LIGenome_Assemblies reconstructed with ONT reads and polished with ONT reads (ONT-only assemblies); (strain_name_ _ONT.fasta) C_LIO_LIGenBank flatfiles from Hybrid assemblies; (strain_name_ _Hyb.gbk) C_LIO_LIGenbank flatfiles from ONT-only assemblies. (strain_name_ _ONT.gbk) C_LI C_LIO_LIData_Tables and Python_code for modular analysis: DOI: doi.org/10.5281/zenodo.15579791 O_LI9728_10237_ORA+HRA_all_reps_microbemod_output.zip C_LIO_LIMethylation calls mapped to parsed genbank features & feature count matrix files. - [Formula] 9728_HRA_reproducibility_analysis_tables.zip - [Formula] 9728_ORA_reproducibility_analysis_tables.zip - [Formula] 10237_HRA_5mC_6mA_reproducibility_analysis_tables.zip - [Formula] 10237_ORA_reproducibility_analysis_tables.zip C_LIO_LIPython code for annotation feature-based modular analysis - [Formula] methylation_feature_hash_assert_FILTER_9728.py - [Formula] methylation_feature_hash_NOFILTER_9728.py - [Formula] methylation_feature_hash_motif_10237.py - [Formula] hyb_vs_ont_per_replicate_comparison.py - [Formula] hyb_vs_ont_combined_plot.py C_LI C_LIO_LIGenome Assembly_Code Availibility: https://github.com/TSchababerle/Bacterial-Methylation O_LIBash script for automated Hybrid Genome Assembly C_LIO_LIBash Script for automated ONT-Only Assembly C_LI C_LI Impact StatementThis study provides a focused proof-of-principle demonstrating the robustness and reproducibility of Oxford Nanopore Technologies (ONT) sequencing-based methylome profiling without short-read correction. Using independent replicates prepared by multiple operators, we show that nanopore-only methods yield consistent, accurate methylation profiles concordant with short-read corrected ONT reads. This highlights the potential of ONT-only sequencing as a broadly accessible, reliable approach for rapid bacterial epigenomic characterization across diverse clinical contexts.

genomics↗

bioRxiv2025

DynamoSort: Using machine learning approaches for the automatic classification of seizure dynamotypes

ObjectiveEpilepsy is characterised by unprovoked and recurring seizures, which can be electrically measured using electroencephalograms (EEG). To better understand the underlying mechanisms of seizures, researchers are exploring their temporal dynamics through the lens of dynamical systems modelling. Seizure initiation and termination patterns of spiking amplitude and frequency can be sorted into "dynamotypes", which may be able to serve as biomarkers for intervention. However, manual classification of these dynamotypes requires trained raters and is prone to variability. To address this, we developed DynamoSort, a machine-learning algorithm for automatic seizure onset and offset classification. MethodsWe used approximately 2100 seizures from an intra-amygdala kainic acid (IAK) mouse model of mesial temporal lobe epilepsy, categorized by five trained raters. MATLABs classification learner application was used to create an ensemble model to score and label dynamotypes of individual seizures based on spiking and frequency features. ResultsDynamotype classification of real EEG data lacks a definitive ground truth, with mean inter-rater agreement at 73.4% for onset and 64.2% for offset types. Despite this, DynamoSort achieved a mean area under the curve (AUC) of 0.81 for onset and a mean AUC of 0.75 for offset types. Machine-human agreement was not significantly different from human-to-human agreement. To address the lack of ground truth in ratings, DynamoSort assigns probabilistic scores (-20 to 20), to indicate similarity to each seizure dynamotype based on spiking features, allowing for a characterization of seizure dynamics on a spectrum rather than the traditional qualitative taxonomy. SignificanceAutomating the classification of dynamotypes is a critical step for their inclusion as a biomarker in clinical and research applications. DynamoSort is a straightforward, open-access tool that uses automatic labelling and probabilistic scoring to quantify subtle changes in seizure onset and offset dynamics. Key PointsO_LIDynamotypes are a promising seizure categorization system, but is prone to interrater variability and lacks a ground truth. C_LIO_LIMachine learning can be used to automatically classify seizure onsets and offsets into appropriate dynamotypes based on spike features. C_LIO_LIAgreement between DynamoSort and human raters rivals typical agreement rates in trained human raters. C_LIO_LIDynamoSort uses probabilistic scoring to quantify subtle changes in seizure onset and offset, allowing for a quantitative characterisation. C_LI

neuroscience↗

bioRxiv2023

Measuring chondrocyte viability of articular cartilage based on label-free two-photon microscopy and deep learning image analysis

ObjectiveChondrocyte viability (CV) is an important indicator of articular cartilage health. Two-photon excitation autofluorescence (TPAF) and second harmonic generation (SHG) microscopy provide a label-free method for imaging chondrocytes. In this study, we propose an automated assessment of CV using deep learning cell segmentation and counting based on acquired TPAF/SHG images. DesignLabel-free TPAF/SHG images of cartilage samples from rats and porcine were acquired using both commercial and home-built two-photon microscopes, respectively. TPAF/SHG images were merged to form RGB color images with red, green, and blue channels assigned to TPAF (two channels) and SHG signals, respectively. To make the training datasets for the deep learning networks, individual chondrocyte areas on the RGB color images were manually circled and live or dead chondrocytes were validated by using Calcein-AM and Ethidium homodimer-1 dye labeling. We first built a chondrocyte viability network (MCV-Net) using the Mask R-CNN architecture, which could provide individual segmented cellular areas with live or dead status. Wiener deconvolution preprocessing was added before the input of MCV-Net to improve the accuracy of the CV analysis, forming the Wiener deconvolution CV network (wMCV-Net). ResultsTraining (300 images) and test (120 images) datasets were built for rats and porcine cartilage respectively. Wiener deconvolution could improve the Peak Signal-to-Noise Ratio (PSNR) for 30-40%. We demonstrated that both MCV-Net and wMCV-Net significantly improved the accuracy of the CV measurement. ConclusionA custom desktop TPAF/SHG microscope was used in collaboration with deep learning algorithm wMCV-Net based label-free method to assess the CV and get 95% accuracy with both rats and porcine samples.

bioengineering↗

bioRxiv2022

Genome editing of the vermilion locus generates a visible eye color marker for Oncopeltus fasciatus

Insects display a vast array of eye and body colors. Genes encoding products involved in biosynthesis and deposition of pigments are ideal genetic markers, contributing, for example, to the power of Drosophila genetics. Oncopeltus fasciatus is an emerging model for hemimetabolous insects, a member of the piercing-sucking feeding order Hemiptera, that includes pests and disease vectors. To identify candidate visible markers for O. fasciatus, we used parental and nymphal RNAi to identify genes that altered eye or body color while having no deleterious effects on viability. We selected Of-vermilion for CRISPR/Cas9 genome editing, generating three independent loss-of-function mutant lines. These studies mapped Of-vermilion to the X-chromosome, the first assignment of a gene to a chromosome in this species. Of-vermilion homozygotes have bright red, rather than black, eyes and are fully viable and fertile. We used these mutants to verify a role for Of-xdh1, ortholog of Drosophila rosy, in contributing to red pigmentation after RNAi. Rather than wild-type-like red bodies, bugs lacking both vermilion and xdh1 have bright yellow bodies, suggesting that ommochromes and pteridines contribute to O. fasciatus body color. Our studies generated the first gene-based visible marker for O. fasciatus and expanded the genetic toolkit for this model system.

genetics↗

bioRxiv2020

Glycopeptidomics Analysis of a Cell Line Model Revealing Pathogenesis and Potential Marker Molecules for the Early Diagnosis of Gastric MALT Lymphoma

BackgroundGastric mucosa-associated lymphoma (GML) is a mature B cell tumor related to Helicobacter pylori (H.pylori) infection. The clinical manifestations of GML are not specific, so GML commonly escapes diagnosis or is misdiagnosed, leading to excessive treatment. The pathogenesis of H.pylori-induced GML is not well understood and there are no molecular markers for early GML diagnosis. MethodsGlycopeptidomics analyses of host cell lines (a BCG823 cell line, C823) and C823 cells infected by H. pylori isolated from patients with GML (GMALT823), gastritis (GAT823), gastric ulcer (GAU823) and gastric cancer (GAC823) were carried out to clarify the host reaction mechanism against GML and identify potential molecular criteria for the early diagnosis of GML. FindingsThirty-three samples were analyzed and approximately 2000 proteins, 200 glycoproteins and 500 glycopeptides were detected in each sample. O-glycans were the dominant glycoforms in GMALT823 cells only. Four specific glycoforms in GMALT823 cells and 2 specific glycoforms in C823 and GMALT823 cells were identified. Eight specific glycopeptides of from 7 glycoproteins were found in GMALT823 cells; of these glycopeptides, 6 and 3 specific glycopeptides had high affinity for T cell epitopes and have conformational B cell epitopes, respectively. InterpretationThe relationship between the predominant glycoforms of host cells and the development of host disease was determined, and the glycoproteins, glycosylation sites and glycoforms might be closely related to the formation of GML, which provides new insight into the pathogenic mechanisms of H. pylori infection and suggests molecular indicators for the early diagnosis of GML.

microbiology↗

bioRxiv2019

Direct readout of neural stem cell transgenesis with an integration-coupled gene expression switch

SUMMARYStable genomic integration of exogenous transgenes is critical for neurodevelopmental and neural stem cell studies. Despite the emergence of tools driving genomic insertion at high rates with DNA vectors, transgenesis procedures remain fundamentally hindered by the impossibility to distinguish integrated transgenes from residual episomes. Here, we introduce a novel genetic switch termed iOn that triggers gene expression upon insertion in the host genome, enabling simple, rapid and faithful identification of integration events following transfection with naked plasmids accepting large cargoes. In vitro, iOn permits rapid drug-free stable transgenesis of mouse and human pluripotent stem cells with multiple vectors. In vivo, we demonstrate accurate cell lineage tracing, assessment of regulatory elements and mosaic analysis of gene function in somatic transgenesis experiments that reveal new aspects of neural progenitor potentialities and interactions. These results establish iOn as an efficient and widely applicable strategy to report transgenesis and accelerate genetic engineering in cultured systems and model organisms.

developmental biology↗

bioRxiv2019

Identifying conservation priorities in a defaunated tropical biodiversity hotspot

AimUnsustainable hunting is leading to widespread defaunation across the tropics. To mitigate against this threat with limited conservation resources, stakeholders must make decisions on where to focus anti-poaching activities. Identifying priority areas in a robust way allows decision-makers to target areas of conservation importance, therefore maximizing the impact of conservation interventions.\n\nLocationAnnamite mountains, Vietnam and Laos.\n\nMethodsWe conducted systematic landscape-scale surveys across five study sites (four protected areas, one unprotected area) using camera-trapping and leech-derived environmental DNA. We analyzed detections within a Bayesian multi-species occupancy framework to evaluate species responses to environmental and anthropogenic influences. Species responses were then used to predict occurrence to unsampled regions. We used predicted species richness maps and occurrence of endemic species to identify areas of conservation importance for targeted conservation interventions.\n\nResultsAnalyses showed that habitat-based covariates were uninformative. Our final model therefore incorporated three anthropogenic covariates as well as elevation, which reflects both ecological and anthropogenic factors. Conservation-priority species tended to found in areas that are more remote now or have been less accessible in the past, and at higher elevations. Predicted species richness was low and broadly similar across the sites, but slightly higher in the more remote site. Occupancy of the three endemic species showed a similar trend.\n\nMain conclusionIdentifying spatial patterns of biodiversity in heavily-defaunated landscapes may require novel methodological and analytical approaches. Our results indicate to build robust prediction maps it is beneficial to sample over large spatial scales, use multiple detection methods to increase detections for rare species, include anthropogenic covariates that capture different aspects of hunting pressure, and analyze data within a Bayesian multi-species framework. Our models further suggest that more remote areas should be prioritized for anti-poaching efforts to prevent the loss of rare and endemic species.

ecology↗

bioRxiv2019

Scalable probabilistic PCA for large-scale genetic variation data

Principal component analysis (PCA) is a key tool for understanding population structure and controlling for population stratification in genome-wide association studies (GWAS). With the advent of large-scale datasets of genetic variation, there is a need for methods that can compute principal components (PCs) with scalable computational and memory requirements. We present ProPCA, a highly scalable method based on a probabilistic generative model, which computes the top PCs on genetic variation data efficiently. We applied ProPCA to compute the top five PCs on genotype data from the UK Biobank, consisting of 488,363 individuals and 146,671 SNPs, in less than thirty minutes. Leveraging the population structure inferred by ProPCA within the White British individuals in the UK Biobank, we scanned for SNPs that are not well-explained by the PCs to identify several novel genome-wide signals of recent putative selection including missense mutations in RPGRIP1L and TLR4.\n\nAuthor SummaryPrincipal component analysis is a commonly used technique for understanding population structure and genetic variation. With the advent of large-scale datasets that contain the genetic information of hundreds of thousands of individuals, there is a need for methods that can compute principal components (PCs) with scalable computational and memory requirements. In this study, we present ProPCA, a highly scalable statistical method to compute genetic PCs efficiently. We systematically evaluate the accuracy and robustness of our method on large-scale simulated data and apply it to the UK Biobank. Leveraging the population structure inferred by ProPCA within the White British individuals in the UK Biobank, we identify several novel signals of putative recent selection.

bioinformatics↗