SEARCH · bioRxiv Science

Results for “molecular biology”

Search indexed bioRxiv preprints in genomics, neuroscience, cell biology and bioinformatics. Read source abstracts and check manuscript versions; preprints are not peer reviewed.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 883 records · Page 49Linked to original sources

bioRxiv2026

A stable cryogenic fluorescence microscope for correlative super-resolution light and electron microscopy

Cryogenic correlative light and electron microscopy (cryo-CLEM) enables visualization of biological specimens with molecular specificity while preserving near-native macromolecular structure. However, the severely limited resolution of conventional cryo-fluorescence microscopes restricts the accuracy of correlation with cryo-electron microscopy. Super-resolution cryogenic CLEM (SR-cryo-CLEM) offers a potential solution, but presents substantial technical challenges, including mechanical instability and ice contamination. Here, we introduce a modular cryogenic light microscope optimized for single-molecule localization microscopy (cryo-SMLM) that mitigates such limitations. The system is constructed primarily from off-the-shelf components, enabling straightforward and cost-effective assembly, and is operated using fully open-source Python software for flexible and customizable control. The mechanically and thermally stabilized architecture, combined with an axial focus-lock system, maintains sample positioning within a standard deviation of 40 nm. Ice contamination is minimized by imaging inside a purged enclosure, enabling prolonged acquisitions. Together, the platform provides robust localization precision, reproducible imaging performance, and an accessible solution for SR-cryo-CLEM.

molecular biology↗

bioRxiv2026

BioMADE: Predicting Torsades de Pointes from molecular structures through biologically informed representations

Drug-induced arrhythmias, particularly Torsades de Pointes (TdP), pose a significant risk to patient safety and can sometimes have life-threatening outcomes. They remain a major concern in drug development and regulation. Machine learning (ML) has become a powerful tool for analyzing complex biological and chemical datasets, enabling researchers to identify subtle patterns that differentiate safe compounds from those likely to cause dangerous cardiac effects. However, most existing in silico approaches do not sufficiently incorporate biological elements, relying heavily on chemical and structural properties or on computationally expensive simulations. Here, we introduce BioMADE, a novel ML framework that harnesses small-molecule-protein activity profiles from publicly available datasets to predict TdP risk without requiring exhaustive mechanistic annotation. Activity data from ChEMBL were used to train individual models for each gene, which predict activity values for any given compound. A curated set of arrhythmia-relevant genes was then used to construct a latent biological embedding (BioMADE embedding) for each molecule. We validated the performance of these features in distinguishing biological elements such as ATC3 class, showing superior classification performance compared with representations such as Molformer (lacks biological information) and MACCS (limited chemical properties) (0.85 AUROC vs 0.81 and 0.73, respectively). BioMADE representations served as input to a support vector machine classifier to discriminate TdP-inducing drugs from safe compounds. BioMADE achieved an AUROC of 0.89 in internal validation, indicating strong predictive performance. Against state-of-the-art models such as ADMEThyst, BioMADE achieved an AUROC of 0.74 on ADMEThysts validation set (vs. 0.72 for ADMEThyst). When we combined both approaches, the AUROC reached 0.77. These results demonstrate that BioMADE provides a scalable, biology-informed, and generalizable approach for predicting drug-induced toxicities. By integrating protein activity profiles into toxicology modeling, our framework highlights the critical role of human biology in adverse drug reaction prediction, an aspect often overshadowed by purely chemical or structural descriptors.

bioinformatics↗

bioRxiv2022

Integrating Self-Attention Transformer with Triplet Neural Networks for Protein Gene Ontology Prediction

Accurate identification of protein function is critical to elucidate life mechanism and design new drugs. We proposed a novel deep-learning method, ATGO, to predict Gene Ontology (GO) attributes of proteins through a triplet neural-network architecture embedded with pre-trained self-attention transformer models. The method was systematically tested on 1068 non-redundant benchmarking proteins and 3328 targets from the third Critical Assessment of Protein Function Annotation (CAFA) challenge. Experimental results showed that ATGO achieved a significant increase of the GO prediction accuracy compared to the state-of-the-art approaches in all aspects of molecular function, biological process, and cellular component. Detailed data analyses showed that the major advantage of ATGO lies in the utilization of attention transformer models which can extract discriminative functional pattern from the feature embeddings. Meanwhile, the proposed triplet network helps enhance the association of functional similarity with feature similarity in the sequence embedding space. In addition, it was found that the combination of the network scores with the complementary homology-based inferences could further improve the accuracy and coverage of the predicted models. These results demonstrated a new avenue for high-accuracy deep-learning function prediction that is applicable to large-scale protein function annotations from sequence alone. AvailabilityThe benchmark dataset, standalone package, and online server for ATGO are available at https://zhanggroup.org/ATGO/. Author SummaryIn the post-genome sequencing era, a major challenge in computational molecular biology is to annotate the biological functions of all gene and gene products, which have been classified, in the context of the widely used Gene Ontology (GO), into three aspects of molecular function, biological process, and cellular component. In this work, we proposed a new open-source deep-learning architecture, ATGO, to deduce GO terms of proteins from the primary amino acid sequence, through the integration of the triplet neural-network with attention transformer models. Large benchmark tests showed that, when powered with a pre-trained self-attention transformer model, ATGO achieved a significantly improved performance than other state-of-the-art approaches in all the GO aspect predictions. Following the rapid progress of self-attention neural network techniques, which have demonstrated remarkable impacts on language processing and multi-sensory data process, and most recently on protein structure prediction, this study showed the significant potential of attention transformer models on protein function annotations.

bioinformatics↗

bioRxiv2022

Comparative Study of State-Dependent Cholesterol Binding Sites in Adenosine A2A and A1 Receptors Using Coarse-Grained Molecular Dynamics Simulations in Biologically Relevant Membranes

In this study we used coarse-grained molecular dynamics (CG MD) simulations to study protein-cholesterol interactions for different activation states of the A2A adenosine receptor (A2AR) and A1 adenosine receptor (A1R). Three different membranes were used, of which a plasma mimetic membrane best matched cholesterol binding sites previously detected in X-ray structures and computationally for the inactive state of A2AR. Here, cholesterol binds stably between TM5 and TM6 in the intracellular leaf and between the extracellular part of TM6 and TM7. The stability of the identified binding sites to A1R with CG MD simulations were further investigated using potential of mean force calculations combined with umbrella sampling. Cholesterol binding to certain cavities in active or inactive state can stabilize a conformation of active or inactive state respectively and is important to identify these cavities. For the active state the cholesterol binding sites have much longer residence times compared to the inactive state for both A2AR and A1R. We observed for the active A1R two cholesterol binding sites in the extracellular membrane leaf in a cavity between TM2/TM3 and along TM3 in the intracellular leaf. We found a partially overlapped binding area between TM6 and TM7 in the extracellular membrane leaf for active state and inactive A1R state which can be antagonized by cholesterol molecules although the site has a much longer residence time for active A1R. For the active A2AR we predicted a high residence time cholesterol position between TM1 and TM7 in the middle region of TMD which is different from the binding sites calculated in active or inactive A1R and for the inactive A2AR. For the inactive A2AR cholesterol binds to the same extracellular area between TM6 and TM7 as in the inactive A1R. The differences in stable, long residence time cholesterol binding sites between active and inactive states of A1R and for A2AR can be important for functional activity and orthosteric agonist or antagonist affinity and can be used for the design of allosteric modulators which can bind through lipid pathways.

biophysics↗

bioRxiv2021

Adaptive ratchets and the evolution of molecular complexity

Biological systems have evolved to amazingly complex states, yet we do not understand in general how evolution operates to generate increasing genetic and functional complexity. Molecular recognition sites are short genome segments or peptides binding a cognate recognition target of sufficient sequence similarity. Such sites are simple, ubiquitous modules of sequence information, cellular function, and evolution. Here we show that recognition sites, if coupled to a time-dependent target, can rapidly evolve to complex states with larger code length and smaller coding density than sites recognising a static target. The underlying fitness model contains selection for recognition, which depends on the sequence similarity between site and target, and a uniform cost per unit of code length. Site sequences are shown to evolve in a specific adaptive ratchet, which produces selection of different strength for code extensions and compressions. Ratchet evolution increases the adaptive width of evolved sites, accelerating the adaptation to moving targets and facilitating refinement and innovation of recognition functions. We apply these results to the recognition of fast-evolving antigens by the human immune system. Our analysis shows how molecular complexity can evolve as a collateral to selection for function in a dynamic environment.

biophysics↗

bioRxiv2022

The cancer-associated RBM39 bridges the pre-mRNA, U1 and U2 snRNPs to regulate alternative splicing

Pharmacologic depletion of RNA-binding motif 39 (RBM39) using aryl sulfonamides represents a promising anti-cancer therapy. However, its efficiency correlates with the expression level of DCAF15 which acts at the interface between RBM39, the drug and the E3-ubiquitin ligase. Consequently, the identification of alternative approaches to deplete RBM39 independently of DCAF15 is required. Here, we combined transcriptomic analysis, functional assays, and structural biology to elucidate the molecular mechanisms governing RBM39 homeostasis. Our data revealed that RBM39 autoregulates the splicing of its own pre-mRNA by triggering the inclusion of a poison exon. During this process, RBM39 selects the 3-splice site of the toxic exon, helps the recruitment of U1 snRNP on its weak 5-splice site and bridges the 3-splice site recognition machinery. The elucidation of the molecular mechanisms controlling RBM39 homeostasis provides unprecedented insights into alternative 3-splice site selection and a solid frame to design alternative anti-cancer therapies.

molecular biology↗

bioRxiv2023

Syntaxin of plants 32 regulates pollen wall development and pollen tube cell wall integrity via controlling secretory pathway

Pollen tubes (PTs) elongate in a polar way to deliver sperm cells to the ovule. Pollen wall development and PT cell wall integrity (CWI) maintenance are critical for PT growth and double fertilization. Pollen wall development mainly relies on secretion of exine precursors in tapetum. RALF4/19-ANX/BUPS-MRI and RALF4/19-LRX-AUN are two distinct signaling pathways but converge to fine-tune CWI during PT growth. Here, we discovered that atsyp32+/-, AtSYP32 RNAi and AtSYP3132 RNAi lines were male sterile. The tapetum development in these lines were disturbed, and the pollen wall structure was impaired resulting in pollen grain and tube bursting and less PTs navigated to micropyles. Strikingly, there were numerous ectopic secretory vesicles retained in pollen cytoplasm, and the abundance or distribution of polysaccharides and AGPs altered significantly in PTs of the mutants and RNAi lines. AtSYP32 interacted with the vesicle transport regulators SEC31B, SEC22 and BET12, the PT CWI regulators RALF19 and LRX11, and the XyG xylosyltransferase XXT5, in the Golgi apparatus. Transcription of some genes related to pollen wall biosynthesis and PT CWI maintenance were seriously affected by AtSYP32 downregulation. Our findings illustrate that AtSYP32 plays essential roles in pollen wall development and PT CWI maintenance via controlling secretory pathway. IN A NUTSHELLO_ST_ABSBackgroundC_ST_ABSPollen wall is the most complex cell wall. Pollen wall development mainly relies on secretion of precursors of exine and pollen coat in tapetal cells. Pollen tubes (PTs) grow in a polar way to deliver sperm cells to the ovule. Maintenance of PT cell wall integrity (CWI) is critical for PT elongation and double fertilization. RALF4/19 ligands interact with BUPS-ANX receptors, signaling it in an autocrine manner to maintain CWI during PT elongation. RALF4/19-LRX-AUN pathway is distinct with RALF4/19-ANX/BUPS-MRI pathway but they converge to fine-tune CWI during PT growth. Biosynthesis of PT cell wall involves multiple subcellular compartments and vesicle transport pathways. Golgi apparatus acts as a hub in vesicle trafficking. Golgi-syntaxin AtSYP31 and AtSYP32 regulate pollen development by controlling intra-Golgi transport and Golgi morphology QuestionWhat is AtSYP32 role in pollen wall and tapetum development? Who are the AtSYP32 partners that regulate secretion of cell wall biosynthesis materials? FindingsWe found that no homozygote progeny was obtained from self-pollinated atsyp32+/- alleles due to pollen sterile. The tapetum development and degeneration in atsyp32+/- mutants was severely delayed, and the pollen wall and PT wall structure were impaired. Strikingly, there were numerous ectopic secretory vesicles retained in pollen cytoplasm in atsyp32+/- mutants, and the abundance or distribution of PT wall polysaccharides and AGPs altered obviously. AtSYP32 interacted with the vesicle transport regulators SEC31B, SEC22 and BET12, the PT CWI regulators RALF19 and LRX11, and XyG xylosyltransferase XXT5, in the Golgi. All these highlight that AtSYP32 regulates pollen wall development and maintenance of PT CWI via controlling secretory pathway. Next stepsThe biological significances and the molecular mechanisms of AtSYP32 interacting with XXT5, RALF19 and LRX11 are elusive but thought-provoking. We are going to clarify the mechanisms.

molecular biology↗

bioRxiv2019

Family saga: reconstructing the evolutionary history of a functionally diverse gene family reveals complexity at the genetic origins of novelty

Gene duplication is associated with the evolution of many novel biological functions at the molecular level. The dominant view, often referred to as "neofunctionalization", states that duplications precede many novel gene functions by creating functionally redundant copies which are less constrained than singletons. However, numerous alternative models have been formulated, including some in which novel functions emerge prior to duplication. Unfortunately, few studies have reconstructed the evolutionary history of a functionally diverse gene family sufficiently well to differentiate between these models. Here we examined the evolution of the g2 family of phospholipase A2 (EC 3.1.1.4) in the genomes of 93 species from all major lineages of Vertebrata. This family is evolutionarily important and has been co-opted for a diverse range of functions, including innate immunity and venom. The genomic region in which this family is located is remarkably syntenic. This allowed us to reconstruct all duplication events over hundreds of millions of years of evolutionary history using manual annotation of gene clusters, which enabled the discovery of a large number of previously un-annotated genes. Intriguingly, we found that the same ancestral gene in the phospholipase gene cluster independently acquired novel molecular functions in birds, mammals and snake, and all subsequent expansion of the cluster originates from this locus. This suggests that the locus has a deep ancestral propensity for multiplication, likely conferred by a structural arrangement of genomic material (i.e. the "genomic context" of the locus) that dates back at least the amniote MRCA. These results highlight the underlying complexity of gene family evolution, as well as the historical- and context-dependence of gene family evolution. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=137 SRC="FIGDIR/small/583344v3_ufig1.gif" ALT="Figure 1"> View larger version (33K): org.highwire.dtl.DTLVardef@1c3228forg.highwire.dtl.DTLVardef@12053forg.highwire.dtl.DTLVardef@11683c0org.highwire.dtl.DTLVardef@123eb3a_HPS_FORMAT_FIGEXP M_FIG C_FIG

molecular biology↗

bioRxiv2020

Global distribution patterns of marine nitrogen-fixers by imaging and molecular methods

Biological nitrogen fixation plays a critical role in marine primary production, yet, our understanding of marine N2-fixers (diazotrophs) is hindered by limited observations. Here, we developed a quantitative image analysis pipeline in concert with mapping of molecular markers for mining >2,000,000 images and >1,300 metagenomes in surface, deep chlorophyll maximum and mesopelagic samples across 6 size fractions (<0.2-2000 m). Imaging and PCR-free molecular data were remarkably congruent. Sequences from diazotrophs were detected from the ultrasmall bacterioplankton (<0.2 m) to mesoplankton (180-2000 m), while images predicted symbiotic and colonial-forming diazotrophs (>20 {micro}m). Imaging and molecular data estimated that polyploidy can significantly impact gene abundances of symbiotic vs colonial-forming diazotrophs. In general our results support the canonical view that larger sized diazotrophs (>10 m) dominate the tropical belts, while sequences from unicellular cyanobacterial and non-cyanobacterial diazotrophs were globally distributed in surface and the mesopelagic. Co-occurring diazotrophic lineages of different lifestyles were frequently encountered, and several new high density regions of diazotrophs were identified in the global ocean. Overall, this work provides an update of marine diazotroph biogeographical diversity and contributes a new bio-imaging-informatic workflow.

microbiology↗

bioRxiv2023

The biological activity of bacterial rhamnolipids is linked to their molecular structure

Biosurfactants are amphiphilic compounds of microbial origin with a wide range of industrial applications. Rhamnolipids (RLs) offer a broad range of potential applications as biosurfactants in industry and agriculture. Several studies report a high capacity for controlling different plant pests and pathogens and consider RLs as promising candidates for bio-based plant protection agents. RLs are a class of glycolipids consisting of one (mRLs) or two (dRLs) rhamnose moieties linked to a single or branched {beta}-fatty acid. Due to the diversity of RLs, so far little is known about the relation between molecular structure and biological activity. Engineering the synthesis pathway allowed us to differentiate between the activities of mixtures of pure mRL and of pure dRL congeners and elaborated HPLC techniques further enabled to analyse the activity of single congeners. In a model system with the plant Arabidopsis thaliana and the plant-parasitic nematode Heterodera schachtii we demonstrate that RLs can significantly reduce infection, whereas their impact on the host plant varies depending on their molecular structure. While mRLs reduced plant growth even at low concentration, dRLs showed no or a beneficial impact on plant development. In addition, the effect of RLs on plant H2O2 production was measured as an indicator of plant defense activity. Treatment with mRLs or dRLs at a concentration of 50 ppm increased H2O2 production. Lower concentrations of up to 10 ppm were used to stimulate plants prior to a treatment with water or flagellin (flg22), a bacterial inducer of plant defense responses. At 10 ppm both mRLs and dRLs fostered an increased response to flg22. However, mRLs also led to an increased response to water, the non-inducing negative control, indicating a generally elevated stress level. Neither mRLs nor dRLs induced expression of plant defense marker genes of salicylic acid, jasmonic acid and ethylene pathways within a 1 hour and 48 hours treatment. Due to the negative effect of mRLs on plants further studies were concentrated on dRLs. Treatment of pre-parasitic infective juveniles of H. schachtii revealed that dRLs did not increase mortality even at a very high concentration of 755 ppm. In order to analyse the effect of single dRL congeners nematode infection assays were performed. While dRL congeners with a C10-C8 acyl chain increased nematode infection, dRLs with C10-C12 and C10-C12:1 acyl chains reduced nematode infection even at concentrations below 2 ppm. Plant growth was not reduced by C10-C8 dRLs, but by C10-C12 and C10-C12:1 dRLs at concentrations of 8.3 ppm. H2O2 production was increased compared to the water control upon treatment with C10-C8 dRLs at a concentration of 200 ppm, while C10-C12 and C10-C12:1 dRLs triggered the same effect already at 50 ppm. Our experiments show a clear structure-effect relation. In conclusion, functional assessment and analysis of mode of action of RLs in plants require careful consideration of their molecular structure and composition.

molecular biology↗

bioRxiv2025

Genome sequence assembly and annotation of MATA and MATB strains of Yarrowia lipolytica

Yeast is commonly utilized in molecular and cell biology research, and Yarrowia lipolytica is favored by bio-engineers due to its ability to produce copious amounts of lipids, chemicals, and enzymes for industrial applications. Y. lipolytica is a dimorphic yeast that can proliferate in aerobic and hydrophobic environments conducive to industrial use. However, there is limited knowledge about the basic molecular biology of this yeast, including how the genome is duplicated and how gene silencing occurs. Genome sequences of Y. lipolytica strains have offered insights into this yeast species and have facilitated the development of new industrial applications. Although previous studies have reported the genome sequence of a few Y. lipolytica strains, it is of value to have more precise sequences and annotation, particularly for studies of the biology of this yeast. To further study and characterize the molecular biology of this microorganism, a high-quality reference genome assembly and annotation has been produced for two related Y. lipolytica strains of the opposite mating type, strain E122 (MATA) and 22301-5 (MATB). The combination of short-read and long-read sequencing of genome DNA and short-read and long-read sequencing of transcript cDNAs allowed the genome assembly and a comparison with a distantly related Yarrowia strain.

genomics↗

bioRxiv2021

Evaluation of odorant-binding protein-1 as a molecular marker for identifying biological forms and delimitating sibling species of Anopheles stephensi

BackgroundAnopheles stephensi, an invasive malaria vector, has been reported to have three biological forms identifiable mainly based on the number of ridges present on the eggs floats. Recently, the first intron of the odorant-binding protein-1 (AsteObp1) has been introduced as a molecular marker for the identification of these forms, and based on this marker, the presence of three putative sibling species (designated as species A, B and C) has been proposed. However, there is no data on the association of proposed markers with biological form or putative species on field populations. MethodsField collected and laboratory-reared An. stephensi were characterized for biological forms based on the number of ridges on the egg s float. DNA sequencing of the partial AsteObp1 gene of An. stephensi individuals were performed by Sangers method, either directly or after cloning with a plasmid vector. Additionally, AsteObp1 sequences of various laboratory lines of An. stephensi were retrieved from a public sequence database. ResultsAsteObp1 intron-1 in Indian An. stephensi populations are highly polymorphic with the presence of more than 13 haplotypes exhibiting nucleotides as well as length-polymorphism (90-to-121 bp). No specific haplotype or a group of closely related haplotypes of intron-1 was found associated with any biological form identified morphologically. High heterozygosity for this marker with a low inbreeding coefficient in field and laboratory populations indicates that this marker is not suitable for the delimitation of putative sibling species, at least in Indian populations. ConclusionsAsteObp1 cannot serve as a marker for identifying biological forms of An. stephensi or putative sibling species in Indian populations.

molecular biology↗

bioRxiv2019

The Work Performed By Biological Motors

Molecular motors are enzymes that perform work (F {middle dot} x) when they move along a track a distance x against a constant force F. This work is performed through intermediate chemical steps in a motors ATPase reaction cycle, each step having a free energy change associated with it that is a sum of chemical, {Delta}{micro}chem, and mechanical, {Delta}{micro}ext, potentials. Defining {Delta}{micro}ext is fundamental to our understanding of how molecular motors work, yet after decades of study the definition of {Delta}{micro}ext remains disputed. Some postulate that {Delta}{micro}ext is a function of both F and x, while others assume that {Delta}{micro}ext is a function of neither F nor x, and still others argue that {Delta}{micro}ext is a function of F but not x. Here we evaluate these models and conclude that only the latter - a mechanochemical model proposed by A.V. Hill in the 1930s - describes molecular motor mechanochemistry.

biophysics↗

bioRxiv2017

Coat color allele and mtDNA haplotype distributions in Russian cat populations: a citizen-science research

This work has started as a project for the summer school in molecular and theoretical biology. Efficient teaching and learning modern methods of molecular biology and population genetics in a two-week study course for high school students needs an attractive subject. We chose phylogeography of the domestic cat as the subject of our project because 1) everybody likes cats; 2) cats are polymorphic for several coat color mutations, which can be easily detected by street survey or in the photographs; 3) samples for DNA extraction are easy to collect without posing ethical and biosafety problems; 4) rich background information of geographical distribution of coat color alleles and variation in mtDNA is available; 5) phylogeography of cat population in Russia is poorly studied and therefore new data collected in this large territory may shed a light on the global cat distribution and on the origin of the fancy breeds.\n\nDuring the project students studied coat color and mitotype distribution across Russian random bred cats. The basics of field (observations of natural cat populations, hair collection), formal (inheritance of coat colors), mathematical population (allele frequency distributions and their comparisons, building phylogenetic trees, multidimensional scaling), and molecular (DNA extraction and amplification, quality control of DNA sequencing data) genetics were covered.\n\nWe scored coat color phenotypes in 1182 cats and sequenced mtDNA control region from hair samples of 38 cats from 18 geographical sites. Analysis of coat color alleles frequencies and mitotype distribution confirmed relative homogeneity of gene pools of Russian cat populations, indicating their recent origin. We found several unique mitotypes and demonstrated that OL1 mitotype, previously found only in Siberian fancy breed, was present in random bred cats from several Russian cities. This contributes to the discussion on the origin of the Siberian breed of cats, and supports the view that Siberians is a recent breed created in the 1980s by breeding of selected representatives of the random bred population.

genetics↗

bioRxiv2021

A multimodal study of a first episode psychosis cohort: potential markers of antipsychotic treatment resistance

Treatment resistant (TR) psychosis is considered to be a significant cause of disability and functional impairment. Numerous efforts have been made to identify the clinical predictors of TR. However, the exploration of molecular and biological markers is still at an early stage. To understand the TR condition and identify potential molecular and biological markers, we analyzed demographic information, clinical data, structural brain imaging data, and molecular brain imaging data in 7 Tesla magnetic resonance spectroscopy, from a first episode psychosis cohort that includes 138 patients. Age, gender, race, smoking status, duration of illness, and antipsychotic dosages were controlled in the analyses. We found that TR patients had a younger age at onset, more hospitalizations, more severe negative symptoms, a significant reduction in the volumes of the hippocampus (HP) and superior frontal gyrus (SFG), and a significant reduction in glutathione (GSH) levels in the anterior cingulate cortex (ACC), when compared to non-TR patients. The combination of multiple markers provided a better classification between TR and non-TR patients compared to any individual marker. Our study shows that ACC GSH, HP and SFG volumes, and age at onset could potentially be trait biomarkers for TR diagnosis, while hospitalization and negative symptoms could be used to evaluate the progression of the disease. Multimodal cohorts are essential in obtaining a comprehensive understanding of brain disorders.

neuroscience↗

bioRxiv2026

CryoConvNeXt enables robust identification of low-abundance molecular species in experimental cryo-EM data

Rare but biologically important molecular species are easily lost when reconstruction-based cryo-EM classification is applied to heterogeneous samples dominated by more abundant particles. We developed CryoConvNeXt, a deep-learning classifier that combines cyclic equivariance with adaptive frequency filtering. It is trained on simulated projections and adapted to experimental data by self-training, an unsupervised domain adaptation technique where the model acts as its own annotator. We tested CryoConvNeXt on cryo-EM data collected for this study from controlled binary and ternary mixtures of Catalase, Apoferritin, and HSP60. Manual particle curation provided reference labels for benchmarks with class ratios from 1:1 to 1:16. CryoConvNeXt retained minority-species recall across this range. In contrast, cryoSPARC's recall collapsed in four of six pairwise conditions, as early as 1:4 when Catalase was the minority species. By recovering low-abundance species that conventional classification fails to recover reliably, CryoConvNeXt takes an important step towards solving a central problem in quantitative visual proteomics: measuring the molecular composition of complex biological samples.

biophysics↗

bioRxiv2022

Comparative transcriptome investigation of Nosema ceranae infecting eastern honeybee workers

Apis cerana is the original host for Nosema ceranae, a widespread fungal parasite resulting in bee nosemosis, which leads to severe losses for apiculture industry throughout the world. However, knowledge of N. ceranae infecting eastern honeybees is extremely limited. Currently, the mechanism underlying N. ceranae infection is still largely unknown. Based on our previously gained high-quality transcriptome datasets, comparative transcriptomic investigation was conducted in this work, with a focus on virulence factor-associated differentially expressed genes (DEGs). Microscopic observation showed that A. c. cerana workers midguts were effectively infected after inoculation with clean spores of N. ceranae. Totally, 1411, 604, and 38 DEGs were identified from NcCK vs. NcT1, NcCK vs. NcT2 and NcT1 vs. NcT2 comparison groups. Venn analysis showed that ten up-regulated genes and nine down-regulated ones were shared by aforementioned comparison groups. GO category indicated these DEGs were involved in a series of functional terms relevant to biological process, cellular component, and molecular function, such as metabolic process, cell part, and catalytic activity. Additionally, KEGG pathway analysis suggested that the DEGs were engaged in an array of pathways of great importance, such as metabolic pathway, glycolysis, and biosynthesis of secondary metabolites. Further, expression clustering analysis demonstrated that majority of genes encoding virulence factors such as ricin B lectins and polar tube proteins displayed apparent up-regulation, whereas a few virulence factor-associated genes such as hexokinase gene and 6-phosphofructokinase gene presented down-regulation during the fungal infection. Finally, the expression trend of 14 DEGs was confirmed by RT-qPCR, validating the reliability of our transcriptome datasets. These results together demonstrated that an overall alteration of the transcriptome of N. ceranae occurred during the infection of A. c. ceranae workers, and most of virulence factor-related genes were induced to activation to promote the fungal invasion. Our findings not only lay a foundation for clarifying the molecular mechanism underlying N. ceranae infection of eastern honeybee workers, but also shed light on developing novel targets for microsporidiosis control.

molecular biology↗

bioRxiv2022

SpaGFT is a graph Fourier transform for tissue module identification from spatially resolved transcriptomics

Tissue module (TM) is a spatially organized tissue region and executes specialized biological functions, recurring and varying at different tissue sites. However, the computational identification of TMs poses challenges due to their convoluted biological functions, poorly-defined molecular features, and varying spatially organized patterns. Here, we present a hypothesis-free graph Fourier transform model, SpaGFT, to represent spatially organized features using the Fourier coefficients, leading to an accurate representation of spatially variable genes and proteins and the characterization of TM at a fast computational speed. We implemented sequencing-based and imaging-based spatial transcriptomics, spatial-CITE-seq, and spatial proteomics to identify spatially variable genes and proteins, define TM identities, and infer convoluted functions among TMs in mouse brains and human lymph nodes. We collected a human tonsil sample and performed CODEX to accurately demonstrate molecular and cellular variability within the secondary follicle structure. The superior accuracy, scalability, and interpretability of SpaGFT indicate that it is an effective representation of spatially-resolved omics data and an essential tool for bringing new insights into molecular tissue biology.

bioinformatics↗