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 Read source abstracts and check manuscript versions; preprints are not peer reviewed.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 55 records · Page 3Linked to original sources

bioRxiv2024

Biology exams rarely use visual models to engage higher-order cognitive skills

Visual models are a necessary part of molecular biology education because submicroscopic compounds and processes cannot be directly observed. Accurately interpreting the biological information conveyed by the shapes and symbols in these visual models requires engaging visual literacy skills. For students to develop expertise in molecular biology visual literacy, they need to have structured experiences using and creating visual models, but there is little evidence to gauge how often undergraduate biology students are provided such opportunities. To investigate students visual literacy experiences, we surveyed 66 instructors who taught lower division undergraduate biology courses with a focus on molecular biology concepts. We collected self-reported data about the frequency with which the instructors teach with visual models and we analyzed course exams to determine how instructors incorporated visual models into their assessments. We found that most instructors reported teaching with models in their courses, yet only 16% of exam items in the sample contained a visual model. There was not a statistically significant relationship between instructors self-reported frequency of teaching with models and extent to which their exams contained models, signaling a potential mismatch between teaching and assessment practices. Although exam items containing models have the potential to elicit higher-order cognitive skills through model-based reasoning, we found that when instructors included visual models in their exams the majority of the items only targeted the lower-order cognitive skills of Blooms Taxonomy. Together, our findings highlight that despite the importance of visual models in molecular biology, students may not often have opportunities to demonstrate their understanding of these models on assessments.

scientific communication and education↗

bioRxiv2020

Biological and molecular characterization of six Shiga toxin-producing Escherichiacoli (STEC) strains encoding the stx2 gene in Colombia

Shiga toxin-producing Escherichia coli (STEC) is a bacterial pathogen that cause diarrhea and severe human diseases. Its principal virulence factor are the Shiga toxins Stx1 and Stx2 which have been identified diverse subtypes considered to be responsible for severe complications of STEC infection. These toxins are encoded in temperate bacteriophages and their expression is linked to phage lithic cycle, which is regulated by late genes and the Q anti-terminator protein. The aim of this study was to characterize biologically and molecularly STEC strains encoding stx2 gene isolated from cattle feces in Colombia. We selected six STEC strains, which were evaluated its Stx production, the Stx2 subtypes, induction of the lithic cycle of bacteriophages and its late region. The results evidenced two highlighted strains with high levels of Stx production and induction of the lithic cycle, compared with the others. Likewise, the strains evaluated showed three Stx2 subtypes: Stx2a, Stx2c, and Stx2d. Regarding the late region, most of the strains carried the qO111 allele and only one strain showed differences in the ninG gene. Although the sample was limited, variability was observed in the Stx production assay, induction of the lithic cycle, Stx2 subtypes and late region of the phages, which could indicate the diversity of the phages carrying STEC strains in Colombia.

molecular biology↗

bioRxiv2017

rxncon 2.0: a language for executable molecular systems biology

Large-scale knowledge bases and models become increasingly important to systematise and interpret empirical knowledge on cellular systems. In signalling networks, as opposed to metabolic networks, distinct modifications of and bonds between components combine into very large numbers of possible configurations, or microstates. These are essentially never measured in vivo, making explicit modelling strategies both impractical and problematic. Here, we present rxncon 2.0, the second generation rxncon language, as a tool to define signal transduction networks at the level of empirical data. By expressing both reactions and contingencies (contextual constraints on reactions) in terms of elemental states, both the combinatorial complexity and the discrepancy to empirical data can be minimised. It works as a higher-level language natural to biologists, which can be compiled into a range of graphical formats or executable models. Taken together, the rxncon language combines mechanistic precision with scalability in a composable and compilable language, that is designed for building executable knowledge bases on the molecular biology of signalling systems.

systems biology↗

bioRxiv2025

Squid primary cell culture as a model system and experimental tool

Squid primary cells from various tissues and ages are isolated, maintained in culture, and express exogenous genes. This protocol opens up numerous opportunities in molecular biology, neuroscience, and marine biology, enabling molecular and cellular-level investigations into processes specific to squids, including their complex behaviors, rapid color change mechanisms, RNA editing capabilities, with broader implications in basic cell physiology. We also describe procedures that harness life-stage tractability in Euprymna berryi squids towards future studies of aging in this model marine organism. Highlights- Squid primary cells are isolated from optic lobes, gills, eyes, and the skin - Cells from these tissues can be dissociated from any life stage of the squid - Isolated cells can express exogenous genes through mRNA transfections and are subject to live and/or fixed cell imaging - Specific details for optimized media conditions, trypsinization, plating, and passaging are included

cell biology↗

bioRxiv2022

An electrogenetic toggle switch design

Synthetic biology uses molecular biology to implement genetic circuits that perform computations. These circuits can process inputs and deliver outputs according to predefined rules that are encoded, often entirely, into genetic parts. However, the field has recently begun to focus on using mechanisms beyond the realm of genetic parts for engineering biological circuits. We analyse the use of electrogenic processes for circuit design and present a model for a merged genetic and electrogenetic toggle switch. Computational simulations explore conditions under which bistability emerges in order to identify the circuit design principles for best switch performance. The results provide a basis for the rational design and implementation of hybrid devices that can be measured and controlled both genetically and electronically.

synthetic biology↗

bioRxiv2023

ATOMDANCE: interpretable kernel-based learning for denoising functional and evolutionary comparisons of protein dynamics

Comparative methods in molecular biology and molecular evolution rely exclusively upon the analysis of DNA sequence and protein structure, both static forms of information. However, it is widely accepted that protein function results from changes in dynamic machine-like motions induced by molecular interactions, a type of data for which comparative methods of analysis are challenged by the large fraction of protein motion created by random thermal noise in the surrounding solvent. Here, we introduce ATOMDANCE, a suite of statistical and kernel-based machine learning tools designed for comparing and denoising functional states of protein motion captured in time-series from molecular dynamics simulations. ATOMDANCE employs interpretable Gaussian kernel functions to compute site-wise maximum mean discrepancy (MMD) between learned features of motion representing two functional, malfunctional or evolutionary protein states (e.g. bound vs. unbound, wild-type vs. mutant). ATOMDANCE derives empirical p-values identifying functional similarity/difference in dynamics at each amino acid site on the protein. ATOMDANCE also employs MMD to contextually analyze potential random amino-acid replacements thus allowing for a site-wise test of neutral vs. non-neutral evolution in the divergence of dynamic function in protein homologs. Lastly, ATOMDANCE also employs mixed-model ANOVA combined with graph network community detection to identify functional shifts in protein regions that exhibit time-coordinated dynamics or resonance of motion across sites. Here, we demonstrate the utility of the software for identifying key sites involved in dynamic responses during functional binding interactions involving DNA, small molecule drugs, and virus-host recognition. We also demonstrate its utility in understanding dynamic resonance changes occurring during the allosteric activation of a pathogenic protease. ATOMDANCE offers a user-friendly interface and only requires an input structure, topology and trajectory files for each of the two proteins being compared (i.e .pdb, .prmtop, and .nc). A separate interface for generating molecular dynamics simulations via open-source tools is also offered. Lay Audience Summary - ATOMDANCE is a suite of software pipelines controlled by a single user interface and designed to comprehensively simulate, calculate and compare protein motions between two functional or evolutionary states while controlling for random noise. It is useful for finding amino acid sites on a given protein that are important in binding other proteins, DNA, or drugs/toxins. It can also be used to assess the effect of genetic mutation on protein motion, and identifies regions of sites on proteins that tend to move together or resonate as a whole unit or community.

biophysics↗

bioRxiv2025

A Unified Multi-modal LLM for Dynamic Protein-Ligand Interactions and Generative Molecular Design

BioDynaGen (Biological Dynamics and Generation) is a novel multi-modal framework unifying protein sequences, dynamic binding site conformations, small molecule ligand SMILES, and natural language text into a single discrete token representation. Built upon a general large language model, BioDynaGen employs continuous pre-training and instruction fine-tuning via next-token prediction to address critical gaps in modeling protein dynamics and ligand interactions. This framework enables a diverse range of tasks, including small molecule-protein binding prediction, dynamic pocket design, and ligand-assisted functional generation. By comprehensively integrating these modalities, BioDynaGen offers an advanced framework for understanding and designing complex biological molecular interactions.

bioinformatics↗

bioRxiv2026

Combining amino acid frequency and 1D convolutional neural network embeddings for the identification of protein-protein interactions using a random forest classifier

Predicting protein-protein interactions is a fundamental problem in molecular biology. Experimental approaches for identifying protein-protein interactions are time-consuming and labor-intensive, motivating the development of efficient computational alternatives, including machine learning-based methods. However, conventional machine learning methods often rely on manually engineered features that require substantial domain expertise. In this study, we propose a two-stage framework to address these limitations. In the first stage, a one-dimensional convolutional neural network autoencoder is used to automatically learn latent representations from protein sequences. The quality of these features is evaluated through reconstruction error, reflecting how accurately the model reconstructs the original sequence. In the second stage, these learned features are combined with amino acid frequency-based features to form a hybrid feature set for predicting protein-protein interactions. A systematic comparison is performed between models trained on frequency features alone and those using a hybrid representation. The comparison showed that incorporating one-dimensional convolutional neural network-derived latent features improved the models performance of predicting protein-protein interactions. The dataset was split into training, validation, and test sets. Nested cross-validation was employed, with inner loops for hyperparameter tuning and outer loops for model selection. The random forest classifier achieved the best performance, with a mean receiver operating characteristic-area under curve of 0.91 and a test F1-score of 0.87. These results highlight the effectiveness of integrating deep feature learning with ensemble methods for predicting protein-protein interactions and build upon previous work focused on this fundamental problem. Author SummaryProtein-protein interactions are fundamental in all biological processes. However, predicting these interactions is a key problem in molecular biology. Computational approaches have been tested to address this problem. We applied a mix of machine learning and deep learning to gain insight into the qualities of proteins that engage in interaction. First, we trained a deep learning model, which automatically learned the primary sequence and characters related thereto, reducing bias in the actual prediction process. We combined these features, or latent representations, with amino acid frequency features of protein sequences, and called the two together "hybrid features." Then we performed a systematic comparison of amino acid frequency features-only with hybrid features, among four different machine learning classifiers. Our results suggest that the random forest classifier performed best among all four classifiers at predicting interactions between proteins. We propose that this approach could be used to improve efficiency in testing protein-protein interactions at the bench and may have applications to other biologically relevant molecular interactions.

bioinformatics↗

bioRxiv2024

Elucidating the molecular mechanism of phytochemicals against Parkinson's disease through an integrated systems biology and molecular modeling approach

Withdrawal StatementThe authors have withdrawn their manuscript owing to the inability to complete the revisions or follow up on the manuscript at this time, due to personal circumstances. Therefore, the authors do not wish this work to be cited as reference for the project. If you have any questions, please contact the corresponding author.

bioinformatics↗

bioRxiv2023

A Molecular Basis of Human Brain Connectivity

Neuroimaging is commonly used to infer human brain connectivity, but those measurements are far-removed from the molecular underpinnings at synapses. To uncover the molecular basis of human brain connectivity, we analyzed a unique cohort of 98 individuals who provided neuroimaging and genetic data contemporaneous with dendritic spine morphometric, proteomic, and gene expression data from the superior frontal and inferior temporal gyri. Through cellular contextualization of the molecular data with dendritic spine morphology, we identified hundreds of proteins related to synapses, energy metabolism, and RNA processing that explain between-individual differences in functional connectivity and structural covariation. By integrating data at the genetic, molecular, subcellular, and tissue levels, we bridged the divergent fields of molecular biology and neuroimaging to identify a molecular basis of brain connectivity. One-Sentence SummaryDendritic spine morphometry and synaptic proteins unite the divergent fields of molecular biology and neuroimaging.

neuroscience↗

bioRxiv2024

ema-tool: a Python Library for the Comparative Analysis of Embeddings from Biomedical Foundation Models

Foundation models, which encode patterns in large, high-dimensional data as embeddings, show promise in many machine learning related applications in molecular biology. Embeddings learned by the models provide informative features for downstream prediction tasks, however, the information captured by the model is often not interpretable. One approach to understanding the captured information is through the analysis of their learned embeddings, which in molecular biology so far has mainly focused on visualizing individual embedding spaces. This study introduces a quantitative framework for cross-space comparison, enabling intuitive exploration and comparison of embedding spaces in molecular biology. The framework emphasizes analyzing the distribution of known biological information within embedding space neighborhoods and provides insights into relationships between multiple embedding spaces. Comparison techniques include global pairwise distance measurements as well as local nearest neighbor analyses. By applying our framework to embeddings from protein language models, we demonstrate how embedding space analysis can serve as a valuable pre-filtering step for task-specific supervised machine learning applications and for the recognition of differential patterns in data encoded within and across different embedding spaces. To support a wide usability, we provide a Python library that implements all analysis methods, available at https://github.com/broadinstitute/EmmaEmb.

bioinformatics↗

bioRxiv2020

Temporal stability and genetic diversity of 48-year old T-series phages

T-series phages have been model organisms for molecular biology since the 1940s. Given that these phages have been stocked, distributed, and propagated for decades across the globe, there exists the potential for genetic drift to accumulate between stocks over time. Here we compared the temporal stability and genetic relatedness of laboratory-maintained phage stocks with a T-series collection from 1972. Only the T-even phages produced viable virions. We obtained complete genomes of these T-even phages, along with two contemporary T4 stocks. Performing comparative genomics, we found 12 and 16 nucleotide variations, respectively, in the genomes of T2 and T6; whereas there were ~172 nucleotide variations between T4-sublines when compared with NCBI RefSeq genome. To account for the possibility of artefacts in the NCBI RefSeq, we used the 1972 T4 stock as a reference and compared genetic and phenotypic variations between T4-sublines. Genomic analysis predicted nucleotide variations in genes associated with DNA metabolism and structural proteins. We did not however, observe any differences in growth characteristics or host range between the T4-sublines. Our study highlights the potential for genetic drift between individually maintained T-series phage stocks, yet after 48-years, this has not resulted in phenotypic alternations in these important model organisms. ImportanceT-series bacteriophages have been used throughout the world for various molecular biology researches, which were critical for establishing the fundamentals of molecular biology - from the structure of DNA to advanced gene-editing tools. These model bacteriophages help keep research data consistent and comparable between laboratories. However, we observed genetic variability when compared contemporary sublines of T4-phages to a 48-year old stock of T4. This may have effects in the comparability of results obtained using T4 phage. Here, we highlighted the genomic differences between T4 sublines and examined phenotypic differences in phage replication parameters. We observed limited genomic changes but no phenotypic variations between T4 sublines. Our research highlights the possibility of genetic drift in model bacteriophages.

microbiology↗

bioRxiv2022

Why is the average collateral effect of synonymous mutations so similar across alternative reading frames?

The standard genetic code has been shown to have multiple interesting properties which impact on molecular biology and the evolutionary process. One facet of molecular biology where code structure is particularly important is the origin and evolution of overlapping genes. We have previously reported that the structure of the standard genetic code ensures that synonymous mutations in a protein coding gene will lead to a remarkably similar average "collateral" mutation effect size in at least four out of the five alternative reading frames. Here we show that only 0.26% of alternative codes with the block structure of the standard genetic code perform at least as well as the standard code in this property. Considering this finding within a code optimality framework suggests that this consistent effect size across the different frames may be adaptive. Here we give context for this finding and present a simple model where a trade-off between evolvability and robustness leads to an average mutation effect size which maximises population fitness. This supports the intuition that similar mutation effects across the different alternative reading frames may be an adaptive property of the standard genetic code which facilitates evolvability through the use of alternative reading frames.

evolutionary biology↗

bioRxiv2025

Protein structure prediction and design for high-throughput computing

Recent advances in structural biology and machines learning have resulted in a revolution in molecular biology. This revolution is driven by protein structure prediction and design tools such as Alphafold3, Chai-1, and Boltz-2 which are now able to accurately model protein structures as well as predict protein-complex formation with a variety of substrates at atomic resolution (i.e., DNA, RNA, small ligands, post-translational modifications). The impact of these protein-structure prediction algorithms has been matched by the emergence of in silico protein design platforms (RFdiffusion), which now promise to revolutionize synthetic biology and novel disease therapeutics. Despite their potential to transform molecular biology, the adoption of these algorithms is hindered in part, not only by their high computational requirements, but also by the difficulty in deploying these algorithms on available systems. To help address these barriers, we developed containerized solutions for AlphaFold3, Chai-1, Boltz-2, and RFdiffusion, optimized across a variety of computational architectures (e.g., x86 and ARM). Additionally, we present OmniFold, an optimized wrapper-platform with automatic QC report generation that enables AlphaFold3, Chai-1, and Boltz-2 to perform simultaneously while more efficiently utilizing GPU systems. Precompiled containers and their definition files are available as open source through Sylabs and GitHub. We hope that these containers and repos will help to facilitate reproducibility, accessibility, and accelerate scientific discovery. Availability and implementationSource code for containers is available at: https://github.com/EpiGenomicsCode/ProteinStruct-Containers https://github.com/EpiGenomicsCode/ProteinDesign-Containers https://github.com/EpiGenomicsCode/OmniFold

bioinformatics↗

bioRxiv2025

Hantavirus stability and inactivation

Hantaviruses are zoonotic viruses that can cause highly pathogenic disease, including hantavirus cardiopulmonary syndrome (HCPS) and haemorrhagic fever with renal syndrome (HFRS), in humans with case-fatality rates of up to 50%. However, our understanding of the basic viral life cycle and the underlying causes of viral pathogenesis remains sparse, in large part due to a lack of molecular biology tools for hantaviruses and the need to work in high-containment laboratory facilities with these viruses. Here, we investigated the kinetics of infectious Tula virus (TULV) particle production in Vero E6 cells and subsequent stability in cell culture media. In addition, we evaluated the stability of infectious virus particles in response to different physical and environmental stresses, including heat, freezing, dehydration and UV exposure, answering key questions about the environmental transmission potential of hantaviruses. Interestingly, we observed a remarkable stability of TULV when stored at room temperature or colder, as well as after dehydration, which suggests that hantaviruses could remain infectious for a sustained period of time after being secreted by their host species. Subsequently, we determined the ability of commonly used virus inactivation methods, including RNA and protein extraction buffers, to inactivate TULV both in a cell-free and cell-associated context and found that TULV was efficiently inactivated by all these methods similar to other enveloped RNA viruses. Finally, we successfully validated the complete inactivation using these inactivation methods using the highly pathogenic HCPS-causing New World Andes virus (ANDV) and the HFRS-causing Old World Hantaan virus (HTNV). These results provide valuable information about safe and effective inactivation methods of viral samples and about the environmental risk potential of hantaviruses. Author summaryHantaviruses are ubiquitous rodent viruses and contain some of the most lethal known zoonotic viruses, including Andes virus (ANDV) and Hantaan virus (HTNV), with no FDA-or EMA-approved antiviral treatment or prevention options available. However, studying the molecular biology and pathogenesis of these viruses is significantly impeded by the need for biosafety level (BSL) 3 (or higher) containment facilities for most hantaviruses and a general lack of molecular biology tools. Our study provides a comprehensive analysis of the stability of infectious hantavirus particles in response to different physical and chemical stresses. We demonstrate that a diverse range of hantaviruses are effectively and quickly inactivated by commonly used generic viral inactivation methods. However, we also provide evidence for an inherent stability under environmental conditions that could enable prolonged transmission potential of hantaviruses, even after secretion from their host species. These data are essential information to design safe inactivation methods of infectious hantavirus material and offers insights into the environmental risks of hantavirus infections with implications for laboratory safety and public health measures.

microbiology↗

bioRxiv2025

TrACES of Time: Towards estimating time-of-day of bloodstain deposition by targeted RNA sequencing

BackgroundIn forensic molecular biology, the main task consists of identifying individuals who contributed to biological traces recovered from (potential) crime scenes. However, to support evidence-based reconstruction of the course of activities having taken place at the scene, contextualising information regarding how and when a biological trace was deposited is oftentimes required. ResultsHere we present the development of a forensic molecular biological analysis procedure for the prediction of the time-of-day at which a bloodstain has been deposited by targeted quantification of selected mRNA markers. Time-of-day candidate prediction markers with diurnally rhythmic expression have previously been identified by whole transcriptome sequencing. Here, we build on our previous findings by establishing a targeted cDNA sequencing protocol on an Ion S5 massively parallel sequencing device for the targeted gene expression quantification of 74 time-of-day candidate prediction markers. Based on expression measurements of these markers in 408 blood samples (from 51 individuals deposited at eight time points over a day), we establish and compare different statistical methods to predict time of deposition. The most suitable model employing penalised regression achieved a root mean squared error of 3 hours and 44 minutes with 78 % of predictions being correct within +/- 4 h (evaluated by five-fold cross-validation). ConclusionsOur study provides the first prediction model for time-of-day of bloodstain deposition based on targeted RNA sequencing and thus represents an important step towards forensic trace deposition timing. It thereby relevantly contributes to the growing knowledge on Transcriptomic Analyses for the Contextualisation of Evidential Stains (TrACES).

molecular biology↗

bioRxiv2023

A centrifugation-based method for high-throughput biomaterial separation using magnetic microbeads

Magnetic microbeads are small iron oxide nanoparticles coated with a bioaffinity material that selectively binds to specific biomolecules of interest, enabling their capture and isolation from complex biological samples. Magnetic microbeads are widely used for purification of specific biomolecules in various experiments in molecular biology. However, current methods of manual pipetting to separate supernatants from magnetic microbeads are often inefficient- time-consuming, labor intensive and inaccurate. Furthermore, the use of pipetting robots and liquid handlers specifically designed for multi-well plates can be a cost-prohibitive approach due to the high cost of equipment and disposable supplies. Here, we developed a centrifugation-based method for high-throughput separation of supernatant from magnetic microbeads. To facilitate the centrifugal separation process, we used the 384 transfer plate (Watson, Japan) and a magnetic stand equipped with a 384-well magnetic stand, allowing easy handling of several hundred samples and rapid separation of supernatant from magnetic microbeads. The centrifugal force was used to drive the separation of target molecules from the magnetic microbeads, and sample were successfully separated with relatively high recovery rates. Thus, this technology provides a simple, rapid, and cost- and labor-effective biomolecule separation method with potential applications in various fields, including molecular biology, clinical diagnostics, and biotechnology, and is a valuable addition to the existing toolbox of biomolecule separation methods.

molecular biology↗

bioRxiv2019

An analysis and optimization of growth condition requirements of the fast-growing bacterium Vibrio natriegens

The fast-growing Gram-negative bacterium Vibrio natriegens is an attractive host for a range of applications in molecular biology and biotechnology. Moreover, the remarkable speed of growth of Vibrio natriegens poses fundamental questions on bacterial physiology and metabolism, energy production, DNA replication and protein synthesis, besides others. In order to address such questions, a solid understanding of the physiological and physical/chemical basis of growth requirements is essential. Here we report a systematic analysis of i) various growth media composition, ii) incubation temperature, iii) pH dependence, and iv) salt concentration requirements for optimal growth of V. natriegens strain DSMZ 759. As a result of the studies, the following optimal conditions were Established: LB medium with 2.5 % NaCl, pH 7.0 - 8.5 and incubation at 37{degrees}C under aerobic conditions. Incubation temperatures above 37 {degrees}C slows growth significantly. Incubation temperatures below 37 {degrees}C slows growth, but at a lower rate. Incubation at or below 28 {degrees}C should be avoided. Under such optimized, standard laboratory conditions, a doubling time of td = 13.6 minutes was observed for V. natriegens measured in mid-log growth phase. The optimized conditions presented here for the growth of V. natriegens can be easily applied in any standardly equipped laboratory. For comparison, identical growth conditions for Escherichia coli were analyzed and are presented as well.\n\nIMPORTANCEGoal of this study was to understand the physiological growths requirements of V. natriegens in routine microbiology and molecular biology laboratory settings. The result is a standardized protocol for the optimized growth of the naturally isolated (wild type) V. natriegens strain DSMZ 759. This protocol can be employed for routine application of V. natriegens for any kind of biochemical, molecular biology and genomic studies and utilization under normal laboratory conditions used by many routinely equipped laboratories.

microbiology↗