Biology subjects

Kozlova, E.

Publications and source records attributed to Kozlova, E..

6 recordsLinked to original sources

bioRxiv2025

DLC2Action: A Deep Learning-based Toolbox for Automated Behavior Segmentation

Behavioral analysis is fundamental to neuroscience, yet the manual annotation of actions remains a bottleneck that constrains both the scale and the reproducibility of experiment. Here, we present DLC2Action, an open-source Python toolbox that enables automatic behavior annotation from video, audio and estimated 2D/3D pose tracking data. DLC2Action integrates multiple state-of-the-art deep learning architectures optimized for action segmentation and supports self-supervised learning (SSL) to address annotation scarcity, boosting performance with limited labeled datasets. To enable model comparison, we establish fixed train/test partitions for nine diverse datasets comprising rodent experiments, human cooking studies, and wildlife observation. DLC2Action reached strong performance across those benchmarks. To further showcase the tools versatility, we applied it to Atari gameplay data and found that, in certain games, the players eye movements consistently predict their button presses across subjects. Because DLC2Action features an intuitive graphical user interface (GUI), users can streamline behavior annotation, perform active learning, and assess of model predictions. Diverse pose, video, and annotation formats are supported. Lastly, DLC2Action is modular and thus designed for extensibility, allowing users to integrate new models, dataset features, and methods. The code and benchmarks are available at: https://github.com/amathislab/DLC2action

animal behavior and cognition↗

bioRxiv2025

A novel rat spatial transitive inference paradigm for investigating memory schema and deliberation

Inferential reasoning is a vital cognitive ability that enables animals to navigate novel situations by leveraging existing relational knowledge of memory schema, with hypothesized roles of prefrontal cortical - hippocampal circuits. Transitive inference (TI) tasks test the ability of subjects to infer relationships within a value hierarchy (e.g., A>B>C>D>E) after being trained only on adjacent premise pairs (e.g., A-B, B-C, etc.). In rodents, current TI paradigms are primarily based on odor-cues and have several limitations that preclude investigation of physiological mechanisms underlying schemas and deliberation. To address these challenges, we developed a novel, automated spatial TI task for rats using a radial maze with maze arms as premise elements and a dedicated deliberation zone. Most rats successfully learned the premise pairs over training. Further, animals demonstrated rapid, successful inference (test pair B>D and control pair A>E) within a single test session, with higher initial accuracy than comparable premise pairs, indicating the use of schema-based inference. We also investigated vicarious trial and errors (VTE), a behavioral correlate of spatial deliberation. VTE behavior was elevated on choice trajectories early in learning, when novel premise pairs were introduced, and generally for incorrect trials, corresponding to the hypothesized association of VTEs with uncertainty. Further, rats also exhibited elevated VTE behavior with high variability during inference testing, with individual variability suggestive of varying strengths of schema usage. Our findings demonstrate the feasibility of a rodent spatial TI task that provides new insights into the behavioral correlates of schemas and deliberation for inferential reasoning.

animal behavior and cognition↗

bioRxiv2023

ProteinFlow: a Python Library to Pre-Process Protein Structure Data for Deep Learning Applications

Over the past few years, deep learning tools for protein design have made significant advances in the field of bioengineering, opening up new opportunities for drug discovery, disease prevention or industrial biotechnology. However, despite the growing interest and excitement surrounding these tools, progress in the field is hindered by a lack of standardized datasets for benchmarking. Most models are trained on data from the Protein Data Bank (PDB), the largest repository of experimentally determined biological macromolecular structures. But filtering and processing this data involves many hyperparameter choices that are often not harmonized across the research community. Moreover, the task of splitting protein data into training and validation subsets with minimal data leakage is not trivial and often overlooked. Here we present ProteinFlow, a computational pipeline to pre-process protein sequence and structural data for deep learning applications. The pipeline is fully configurable and allows the extraction of all levels of protein organization (primary to quaternary), allowing end-users to cater the dataset for a multitude of downstream tasks, such as protein sequence design, protein folding modeling or protein-protein interaction prediction. In addition, we curate a feature-rich benchmarking dataset based on the latest annual release of the PDB and a selection of preprocessing parameters that are widely used across the research community. We showcase its utility by benchmarking a state-of-the-art (SOTA) deep learning model for protein sequence design. The open source code is packaged as a python library and can be accessed on https://github.com/adaptyvbio/ProteinFlow.

bioinformatics↗

bioRxiv2020

Unannotated small RNA clusters in circulating extracellular vesicles detect early stage liver cancer

BackgroundHepatocellular carcinoma (HCC) is among the deadliest malignancies and surveillance tools for early detection are suboptimal. Extracellular vesicles (EVs) have gained increasing scientific interest due to their involvement in tumor initiation and metastasis, however, most extracellular RNA (exRNA) biomarker studies are limited to annotated genomic regions. MethodsEVs were isolated with ultracentrifugation and nanoDLD and quality assessed by electron microscopy, immunoblotting, nanoparticle tracking, and deconvolution analysis. We performed genome-wide small exRNA sequencing, including unannotated transcripts. We identified small RNA clusters (smRCs) and delineated their key genomic features across biospecimens (blood, urine, tissue) and EV isolation techniques. A 3-smRC signature for early HCC detection was trained and validated in two independent cohorts. ResultsEV-derived smRCs were dominated by uncharacterized, unannotated small RNA and uniformly tiled across the genome with a consensus sequence of 20bp. A 3-smRC signature was significantly overexpressed in circulating EVs of HCC patients compared to controls at risk or patients with non-HCC malignancies (p<0.01, n=157). An independent validation in a phase 2 biomarker study revealed 86% sensitivity and 91% specificity for the detection of early HCC from controls at risk (i.e. cirrhosis or chronic liver disease, n=209) (positive predictive value (PPV): 89%, area under the ROC curve [AUC]: 0.87). The 3-smRC signature was independent of alpha-fetoprotein (p<0.0001) and a composite model yielded an increased AUC of 0.93 (sensitivity: 85%, specificity: 94%, PPV: 95%). ConclusionAn exRNA-based 3-smRC signature from plasma detects early stage HCC, which directly leads to the prospect of a minimally-invasive, blood-only, operator-independent surveillance biomarker. One sentence summaryWe employ a novel, data-driven approach to identify and characterize small RNA clusters from unannotated loci in extracellular vesicle-derived RNA across different cancer types, isolation techniques, and biofluids, facilitating discovery of a robust biomarker for detection of early stage liver cancer.

cancer biology↗

bioRxiv2019

Mycobacterium tuberculosis hijacks an evolutionary recent IFN-IL-6-CEBP axis linked to monocyte development and disease severity in humans

Monocyte counts are increased during human tuberculosis (TB) but it has not been determined whether Mycobacterium tuberculosis (Mtb) directly regulates myeloid commitment. We demonstrated that exposure to Mtb directs primary human CD34+ cells to differentiate into monocytes/macrophages. In vitro myeloid conversion did not require type I or type II IFN signaling. In contrast, Mtb enhanced IL-6 responses by CD34+ cell cultures and IL-6R neutralization inhibited myeloid differentiation and decreased mycobacterial growth in vitro. Integrated systems biology analysis of transcriptomic, proteomic and genomic data of large data sets of healthy controls and TB patients established the existence of a myeloid IL-6/IL6R/CEBP gene module associated with disease severity. Furthermore, genetic and functional analysis revealed the IL6/IL6R/CEBP gene module has undergone recent evolutionary selection, including Neanderthal introgression and human pathogen adaptation, connected to systemic monocyte counts. These results suggest Mtb co-opts an evolutionary recent IFN-IL6-CEBP feed-forward loop, increasing myeloid differentiation linked to severe TB in humans.

immunology↗

bioRxiv2018

CRISPR/Cas9-mediated gene deletion of the ompA gene in an Enterobacter gut symbiont impairs biofilm formation and reduces gut colonization of Aedes aegypti mosquitoes

BackgroundSymbiotic bacteria are pervasive in mosquitoes and their presence can influence many host phenotypes that affect vectoral capacity. While it is evident that environmental and host genetic factors contribute in shaping the microbiome of mosquitoes, we have a poor understanding regarding how bacterial genetics affects colonization of the mosquito gut. The CRISPR/Cas9 gene editing system is a powerful tool to alter bacterial genomes facilitating investigations into host-microbe interactions but has yet to be applied to insect symbionts. Methodology/Principal FindingsTo investigate the role of bacterial factors in mosquito biology and in colonization of mosquitoes we used CRISPR/Cas9 gene editing system to mutate the outer membrane protein A (ompA) gene of an Enterobacter symbiont isolated from Aedes mosquitoes. The ompA mutant had an impaired ability to form biofilms and poorly infected Ae. aegypti when reared in a mono-association under gnotobiotic conditions. In adults the mutant had a significantly reduced infection prevalence compared to the wild type or complement strains, while no differences in prevalence were seen in larvae, suggesting bacterial genetic factors are particularly important for adult gut colonization. We also used the CRISPR/Cas9 system to integrate genes (antibiotic resistance and fluorescent markers) into these symbionts genome and demonstrated that these genes were functional in vitro and in vivo. Conclusions/SignificanceOur results shed insights onto the role of ompA gene in host-microbe interactions in Ae. aegypti and confirm that CRISPR/Cas9 gene editing can be employed for genetic manipulation of non-model gut microbes. The ability to use this technology for site-specific integration of genes into the symbiont will facilitate the development of paratransgenic control strategies to interfere with arboviral pathogens such Chikungunya, dengue, Zika and Yellow fever viruses transmitted by Aedes mosquitoes. Author summaryMicrobiota profoundly affect their host but few studies have investigated the role of bacterial genetics in host-microbe interactions in mosquitoes. Here we applied the CRISPR/Cas9 gene editing system to knock out a membrane protein in Enterobacter, which is a dominant member of the mosquito microbiome. The mutant strain lacked the capacity to form biofilms, infected larvae and adults at lower titers, and had a reduced prevalence in adults. The lower prevalence in adults, but not larvae, likely reflects the difference in the modes of bacterial acquisition from the larval water of these two life stages. Importantly from an applied perspective, we also demonstrated that this editing technology can be harnessed for site-specific integration of genes into the bacterial chromosome. In proof-of-principle studies we integrated either a fluorescent protein or gene conferring antibiotic resistance into the bacterial genome and showed these transgenes were functional in mosquitoes. The specificity, flexibility, and simplicity of this editing approach in non-model bacteria will be useful for developing novel symbiotic control strategies to control arthropod-borne disease.

microbiology↗