Biology subjects

Mittal, A.

Publications and source records attributed to Mittal, A..

At least 19 recordsLinked to original sources

bioRxiv2026

Reliable and Efficient Agentic Protein Design with Toolset Harnessing

Computational protein design workflows require the orchestration of multiple tools, data formats, and analysis steps that are often fragmented. Here, we introduce Chaperone, a lightweight and customizable framework that enables seamless integration of RFdiffusion3 (RFD3) with downstream sequence design, structure prediction, and evaluation tools, including ProteinMPNN, AlphaFold3 (AF3), and more than 50 other protein design tools. By combining a programmable agent-friendly interface with parallelized execution on high-performance computing systems, Chaperone as a harness enables rapid iteration and scalable exploration of complex protein design workflows. Across an evaluation of nine biomolecular design tasks, we find using Chaperone reduces total token usage relative to baseline agents by 86% and GPU costs by 64%, while increasing the overall computational design pass rate. We demonstrate Chaperones utility for large-scale design-space exploration through a two-week end-to-end small-molecule binder design campaign, yielding low nanomolar binders to cortisol zero shot. We further demonstrate its versatility across distinct design challenges, including the design of a symmetric 12-subunit self-assembly nanoparticle and a linker that mediates interactions between C2- and D3-symmetric protein complexes to form crystalline assemblies. Together, these results demonstrate how agent harnesses built around standardized tool interfaces can substantially improve the efficiency and success of complex, long-horizon agentic design workflows in biology. We make Chaperone freely available to the community on GitHub: https://github.com/RosettaCommons/chaperone.

biochemistry↗

bioRxiv2026

Structural Determinants of Synaptic Vesicle Protein 2C Ligand Selectivity and Their Impact on Dopamine Release

The synaptic vesicle protein 2 (SV2) family is a highly conserved group of transmembrane glycoproteins sharing approximately 70% sequence identity. SV2A and SV2B are broadly distributed throughout the brain, while SV2C, the evolutionarily oldest member of the family, is primarily found in dopaminergic brain regions in the basal ganglia. Genome-wide association studies have linked SV2C to Parkinson's disease, and SV2C appears to enhance dopamine storage in synaptic vesicles, but the basis for this effect is unknown. SV2s are the target of the racetam class of antiseizure medications, and SV2A-specific ligands are widely used to treat seizures. Recently several SV2C-specific ligands have been developed, but the structural basis for ligand specificity is unclear. A better understanding of SV2C structure, function, and pharmacology could lead to better targeted therapies for epilepsy, Parkinson's disease, and other dopamine-related conditions. Here we present cryo-EM structures of apo SV2C, SV2C bound to the high affinity non-selective SV2 ligand padsevonil, SV2C bound to the SV2C-selective ligand UCB-F, and SV2A bound to the SV2A-selective ligand plosaracetam (also known as ABBV-552/SDI-118). We find that SV2C has a wider luminal opening than SV2A and SV2B, which allows for UCB-F to bind to the primary site and form favorable interactions that are not possible in the narrower primary binding site of SV2A and SV2B. We also demonstrate that UCB-F and padsevonil, but not plosaracetam, reduce dopamine release in striatal sections of mouse brain. Our biochemical experiments and structures provide insights into SV2 ligand specificity and offer a template for the rational development of therapeutics targeting SV2C.

biochemistry↗

bioRxiv2026

Discovery of Selective Small-Molecule Ligands of SV2C by AI-Enhanced Virtual Screening and Experimental Validation

Synaptic vesicle glycoprotein 2C (SV2C) is a vesicular protein enriched in dopaminergic neurons of the basal ganglia that modulates dopamine storage and release, and its disruption is implicated in Parkinson's disease (PD). Despite strong genetic and pathological links to PD, there are no selective small-molecule probes for SV2C. Here, we describe an AI-enhanced virtual screening (VS) and experimental campaign that identified multiple novel chemotypes with low-micromolar affinity and marked selectivity for SV2C over SV2A and SV2B, starting from a large, general-purpose commercial library. Because no full-length high-resolution SV2C structure was available, we built a homology model using SV2A cryo-EM structures as templates and characterized its conformational landscape by molecular dynamics (MD) and Gaussian accelerated MD (GaMD) simulations in apo form and in complex with known SV2 ligands (plosaracetam, levetiracetam, brivaracetam, and padsevonil). A convolutional neural network-based scoring function (CNN_VS), retrospectively validated on a manually curated 39-ligand SV2A benchmark (r = 0.72 vs experimental pIC50), was then applied in a multi-stage funnel to 5.96 million Mcule in-stock compounds, which were sequentially filtered to 3.19 million CNS-relevant molecules before docking and rescoring. From 94 VS-prioritized candidates, 71 compounds were experimentally profiled in an orthogonal primary assay cascade combining a thermal shift assay (TSA) with a [3H]-padsevonil scintillation proximity assay (SPA), followed by Ki determination and isoform selectivity profiling for key hits. This campaign yielded 22 active molecules (31% hit rate) that naturally segregated into two categories: compounds that showed primary site competition, and compounds that did not show primary site competition with [3H]-padsevonil. A subset of competitor compounds also showed thermostabilization activity. Among these, compounds 36 and 56 emerged as particularly attractive leads, with Ki values of 24.6 uM and 3.25 uM at SV2C, respectively, and greater than 10-fold selectivity versus SV2A; compound 56 also maintained approximately 12-fold selectivity relative to SV2B. A complementary subset of SV2C-selective hits behaved as padsevonil-site competitors, providing a lead set that will serve as a template for functional characterization and future drug development for conditions that affect dopaminergic signaling. Docking analysis suggests a common binding mode anchored by conserved tryptophan residues in the SV2 pocket, a prediction independently confirmed by an unpublished SV2A-plosaracetam cryo-EM structure showing 0.76 Angstrom binding-site C-alpha RMSD relative to the SV2C model and complete conservation of the tryptophan cage. Subtle differences in the luminal domain and transmembrane region point to the structural determinants underlying isoform selectivity. Collectively, these results demonstrate that an AI-driven VS pipeline, tightly integrated with medium-throughput biophysical assays, can deliver selective SV2C binders from a general chemical library on a structurally under-characterized membrane target. The identified hits provide multiple starting points for hit-to-lead optimization and tools for probing SV2C biology and its role in PD.

neuroscience↗

bioRxiv2026

Killiverse: an interactive multi-omics web resource for killifish

BackgroundKillifish have emerged as valuable vertebrate model systems for investigating several disciplines including aging, regeneration, and developmental biology. Multi-omics datasets are increasingly being generated for killifish. However, their reuse remains limited due to computational challenges, largely due to the lack of accessible resources creating a bottleneck in widespread adoption of the killifish model. To address this, we developed Killiverse, a web resource for quick and intuitive exploration of multi-modal omics data dedicated to the model organism. ResultsKilliverse is an interactive, no-code, web-based platform designed for exploration of killifish multi-omics data. The platform aggregates a growing list of datasets including bulk transcriptomes, single-cell and single-nucleus transcriptomes, proteomes, and lipidomes processed through standardized pipelines and genome assemblies. Killiverse supports customized visualization and enables cross-study and cross-species analysis. It provides ortholog mapping to several established model organisms. By combining low-code software development with modern cloud technologies, the platform delivers a scalable browser-accessible application for the community. ConclusionsKilliverse enables rapid hypothesis development through the identification of patterns across studies and species. The ortholog maps allow the findings to be placed in a broader biological context. The platform represents an innovation in genomics data visualization that will serve as a template for future tool development. Killiverse is freely accessible at https://killiverse.org/.

genomics↗

bioRxiv2026

PAD4-mediated histone citrullination contributes to enhanced NETosis in type 2 diabetes but not in type 1 diabetes

Neutrophil extracellular traps (NETs) are web-like structures released by activated neutrophils through a process known as NETosis, which help immobilize, trap, and eliminate invading pathogens. While NETs play a critical role in host defense, excessive or dysregulated NET formation can contribute to chronic inflammatory diseases such as diabetes and its complications. Elevated levels of NETs and increased expression of its mediator, protein-arginine deiminase type 4 (PAD4), have been reported in diabetes. However, the underlying molecular mechanisms governing the enhanced NETosis in type 1 (T1D) and type 2 diabetes (T2D) remain unclear. Using mouse models and human patient samples, we show that neutrophils undergo enhanced NETosis in T1D and T2D through distinct pathways. We found enhanced NETosis in T1D occurs in a PAD4-independent manner, driven by robust cytosolic ROS production by NADPH oxidase (NOX). This leads to myeloperoxidase activation and chromatin decondensation. We further confirm the PAD4 independent mechanism in neutrophils from STZ-induced PAD4-/- mice. In contrast, in T2D, neutrophils undergo NOX-independent NETosis, which relies on calcium-mediated mitochondrial ROS production, PAD4 activation, and hyper-histone citrullination. Taken together, our findings reveal previously uncharacterized distinct mechanisms of enhanced NETosis in T1D and T2D.

Molecular Biology↗

bioRxiv2026

Diet-conditioned microbiota enhances fecal microbiota transplantation efficacy in alcoholic liver disease through caproic acid-PPARα signaling

Background and AimFecal microbiota transplantation (FMT) in Alcohol-related liver disease (ALD) has shown therapeutic potential, with variable efficacy and unclear mechanism. Because dietary protein influences gut microbiota composition, we hypothesized that donor dietary preconditioning could enhance FMT efficacy. We therefore examined in a murine ALD model if high-protein donor diet improves FMT outcome. MethodsALD was induced in C57BL/6N mice using a Lieber-DeCarli ethanol diet combined with thioacetamide administration for 12 weeks. FMT was performed using stool from diet-modulated donors, and recovery was assessed on day7 post-FMT. Multi-omics analysis using 16s rRNA and mass spectroscopy was performed for Gut microbiota composition, plasma- and stool-metabolome, and hepatic proteomes. Multi-omics outcomes were validated in ALD animal and Huh7 hepatocytes. ResultsBoth protein-based FMTs improved ALD recovery; Veg-FMT demonstrated superior efficacy, significantly reducing hepatic injury (AST 1.2-fold, p=0.002; bilirubin 1.2-fold, p=0.03; steatosis 1.7-fold,p=0.01) and restoring gut barrier integrity (occludin 1.5-fold,p=0.04; mucin 2 2.2-fold, p=002; and plasma endotoxin 1.7-fold, p=0.02). A significant 2-fold increase was observed in Lachnospiraceae NK4A136, Coriobacteriaceae UCG-002, and short-chain fatty acids, particularly caproic acid. Functional validation confirmed that caproic acid promoted hepatic fatty acid {beta}-oxidation through PPAR-dependent mechanisms, reducing triglyceride accumulation and lipogenesis in both cellular and animal models. ConclusionDonor preconditioning with a plant-protein enriched diet enhances FMT efficacy in ALD by gut microbiota modulation with increased metabolites like caproic acid. These findings highlight a microbiota-metabolite-host axis through which diet-modulated FMT improves hepatic lipid metabolism and injury, and identifies a pathway via which FMT imparts its effect. SignificanceThis study identifies a mechanistic basis for improving fecal microbiota transplantation (FMT) efficacy in alcohol-related liver disease (ALD) by demonstrating that dietary preconditioning of donor microbiota improves therapeutic outcomes. We show that plant protein-modulated donor microbiota supplements abstinence-associated recovery through increased production of the microbial metabolite caproic acid, which promotes hepatic fatty acid {beta}-oxidation via PPAR signaling. These findings highlight donor dietary conditioning and microbiota-derived metabolites, rather than microbial composition alone, as important determinants of FMT efficacy. The results suggest that microbial metabolites such as caproic acid may represent potential therapeutic targets or biomarkers to enhance and standardize microbiota-based interventions in ALD. Although the current work is based on a murine model, the identified microbiota-metabolite-host metabolic axis provides a framework for future translational studies aimed at optimizing FMT strategies in liver disease.

microbiology↗

bioRxiv2026

Production of mouse ultrasonic vocalizations and distress calls is associated with different patterns of Fos expression in the nucleus retroambiguus

Animals produce different vocalization types, which differ in their acoustic features and are produced in different behavioral contexts. How vocalization-related brain circuits are organized to enable the production of different vocalization types remains poorly understood. The nucleus retroambiguus is a hindbrain premotor region that regulates the production of both ultrasonic vocalizations (USVs) and distress calls (squeaks) in adult mice, but whether distinct or overlapping populations of RAm neurons are recruited during the production of these two vocalization types is unknown. In the current study, we used Fos immunohistochemistry to compare the counts and spatial distributions of Fos-positive RAm neurons in males and females that produced USVs and females that produced courtship squeaks. We also combined in vivo activity-dependent (TRAP2) labeling with Fos immunohistochemistry to directly compare Fos expression associated with the production of USVs and courtship squeaks in the same females. Our findings suggest that RAm contains three vocalization-related populations of neurons: squeak-related neurons, USV-related neurons, and shared neurons that are recruited during both vocalization types. These findings refine current models of the premotor control of vocalization and set the stage for future work to explore anatomical and functional heterogeneity within RAm.

neuroscience↗

bioRxiv2026

AI/ML-Assisted Computational Design and Immunoinformatics Evaluation of a Multi-Epitope Vaccine Targeting Podoplanin in Glioblastoma Multiforme

Glioblastoma Multiforme (GBM) is one of the most malignant forms of brain tumor in humans, with limited treatment options and poor overall survival rates. In the present study, we employed an in-silico workflow that integrated immunoinformatics and 3D structural modelling tools to design a multi-epitope vaccine against Podoplanin (PDPN), a transmembrane glycoprotein primarily involved in tumor invasion and metastasis. The differential expression of PDPN in tumor versus normal cells was investigated using transcriptomics datasets. Once the overexpression was confirmed, it was designated as a Tumor-Associated Antigen (TAA). B-cell, CTL, and HTL epitopes were predicted and screened for antigenicity, non-allergenicity, and non-toxicity. Selected epitopes were linked with appropriate adjuvant and linker sequences to construct a vaccine candidate. Codon optimization and in silico cloning was conducted to evaluate the constructs expression in a mammalian expression vector. The 3D structure of the vaccine candidate was modelled, refined, and validated before molecular docking with immune receptors and immune simulation studies. The results indicated that proposed polypeptide, RasIC-01v, could be a potential vaccine candidate for highly vigorous and dangerous cancer like GBM. Further experimental and immunological validations would be required to validate the commercial feasibility and development of RasIC-01v. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=116 SRC="FIGDIR/small/706629v1_ufig1.gif" ALT="Figure 1"> View larger version (35K): org.highwire.dtl.DTLVardef@7485b1org.highwire.dtl.DTLVardef@1f551c1org.highwire.dtl.DTLVardef@ca871eorg.highwire.dtl.DTLVardef@6cf53d_HPS_FORMAT_FIGEXP M_FIG C_FIG

immunology↗

bioRxiv2025

Chemical Dice Integrator (CDI): A Scalable Framework for Multimodal Molecular Representation Learning

The machine learning landscape for molecular property prediction is fragmented, with numerous Featurizers each capturing a narrow, specialized view of chemical structure. This heterogeneity forces a suboptimal choice of representation a priori, limiting model generalizability. We introduce the Chemical Dice Integrator (CDI), a hierarchical framework that unifies six orthogonal molecular representations, physicochemical (Mordred), topological (GROVER), visual (ImageMol), biological (Signaturizer), quantum-mechanical (MOPAC), and linguistic (ChemBERTa), into a single, coherent embedding. The framework consists of CDI-Basic, a two-tiered autoencoder that fuses these modalities, and CDI-Generalised, a Mamba State-Space Model (SSM) that learns a direct, efficient map from SMILES strings to the unified embedding space. Extensive benchmarking across 23 classification (171 tasks) and 10 regression datasets demonstrates that CDI embeddings consistently achieve superior predictive performance compared to individual Featurizers and standard feature aggregation methods. The CDI-Generalised model achieves this performance with exceptional computational efficiency, outperforming deep learning Featurizers in terms of speed and resource overhead. Furthermore, we demonstrate that the CDI embedding is chemically intuitive, allowing for the sensitive distinction of nuanced structural variants, such as chiral enantiomers and kekulized SMILES forms. By bridging multimodal chemical intelligence with scalable, sequence-based inference, CDI offers a strong foundation for molecular machine learning.

bioinformatics↗

bioRxiv2025

MutAIverse: An AI-Powered, Mechanism-backed Platform for Discovering Novel DNA Adducts and their precursor Genotoxins

Genotoxin exposure leads to DNA adduct formation, potentially causing mutations if unrepaired. Current DNA adductomics platforms or analytical workflows are limited by incomplete spectral libraries, reliance on experimentally validated adducts, limited cellular contexts, and inefficient computational methodologies. We introduce MutAIverse, an advanced AI-driven DNA adductomics analysis platform that overcomes these limitations by leveraging intracellular mechanistic modeling of genotoxin bioactivation to construct a comprehensive DNA adduct library and offers advanced spectral mapping. MutAIverse integrates experimentally validated and chemically valid putative DNA adducts, enabling interpretable retrograde tracking to parental genotoxins. Validation of MutAIverse against experimental MS/MS datasets demonstrated the detection of both known and novel adducts. Furthermore, application to in-house generated DNA adductomics data from tissue biopsies of smokeless tobacco-induced head and neck cancer patients revealed selective enrichment of both novel and known DNA adducts; a subset of them was validated using MS/MS analysis. Collectively, MutAIverse provides a robust, end-to-end, and interpretable platform for advanced DNA adductomics analysis.

bioinformatics↗

bioRxiv2025

AlignmentFold and AlignmentPartition: Improving the align-then-fold approach for RNA secondary structure prediction

The combination of free energy minimization with sequence covariation analysis is a popular approach for predicting conserved secondary structure using a fixed alignment of homologous RNA sequences. In this work we revisit this approach by developing the new AlignmentFold and AlignmentPartition programs. AlignmentFold and AlignmentPartition predict consensus minimum free energy structures and base pairing probabilities, respectively. We determined new nearest neighbor thermodynamic parameters for gaps and non-canonical base pairs that align with base pairs. These frequently occur for structure prediction with a fixed input alignment. We assessed the impact of alignment prediction method and size on the prediction accuracy. We also assessed the improvement in prediction accuracy conferred by sequence covariation analysis. In contrast to previous solutions, we do not include covariation in the scoring of structure quality and achieve prediction accuracy as good as other tools. AlignmentFold and AlignmentPartition are freely available as part of the RNAstructure software package at https://rna.urmc.rochester.edu/RNAstructure.html.

bioinformatics↗

bioRxiv2025

Mechanisms Underlying Allosteric Modulation of Antiseizure Medication Binding to Synaptic Vesicle Protein 2A (SV2A)

Brivaracetam (BRV) and levetiracetam (LEV) are antiseizure medications (ASMs) that target synaptic vesicle protein 2A (SV2A), while UCB1244283 acts as a positive allosteric modulator of these medications. The SV2A-BRV-UCB1244283 complex reveals how UCB1244283 allosterically enhances BRV binding by occupying an allosteric site near the primary binding site, preventing BRV dissociation. This allosteric site, formed by hydrophobic and uncharged residues, is a novel small-molecule binding site in SV2A. Structural analysis and mutagenesis suggest that an allosteric network between the primary and allosteric sites governs high-affinity ASM binding. UCB1244283 selectively binds SV2A over SV2B and SV2C, with specific mutations disrupting binding. The structure explains why UCB1244283 binding to SV2A selectively allows interaction with specific ASMs but not others due to steric hinderance. Structural comparison reveals that distinct conformational differences between the SV2A-BRV-UCB1244283 and other SV2A-ligand complexes, particularly in the transmembrane domain, influence binding at both sites. Future research will explore potential therapeutics targeting the allosteric site and their impact on SV2A regulation. TeaserUCB1244283 enhances LEV and BRV binding to SV2A, revealing an allosteric site that could aid in developing targeted therapeutics.

biochemistry↗

bioRxiv2025

NF-κB mediates transactivation of HNRNPD, resulting in PTEN destabilization and constitutive activation of the PI3K-AKT pathway in oral cancer cells

Heterogeneous Ribonucleoprotein D (hnRNPD), an RNA binding protein transcriptionally upregulated by NF-{kappa}B transcription factor, is associated with poor outcome of Oral Squamous Cell Carcinoma (OSCC). However, the role of hnRNPD in OSCC remains elusive. This study reveals that hnRNPD positively affects the proliferation, migration, invasion, and survival of OSCC cells. Transcriptome profiling in hnRNPD knockout cells identified significant upregulation of PTEN and inhibition of the PI3K/AKT/mTOR axis. HnRNPD mediates the destabilization of PTEN mRNA by binding to the class II AU-Rich Element (ARE) in 3UTR of PTEN. The expression of hnRNPD and PTEN are strongly negatively correlated in OSCC tissue specimens, further corroborating hnRNPD-mediated PTEN destabilization. The hnRNPD knockout inhibited autophagy, evident by an accumulation of autophagic vesicles and decreased autophagic flux. Mechanistically, the hnRNPD knockout reduced the expression of NF-{kappa}B, eventually downregulating its transcriptional target LC3b, a key mediator of autophagy. SA-{beta}-Galactosidase staining in hnRNPD KO cells conclusively demonstrated the onset of cellular senescence. The present study demonstrates hnRNPD-driven positive modulation of autophagy via NF-{kappa}B, independent of the PI3K/AKT/mTOR axis, highlighting it as a novel therapeutic target for treating oral cancer. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=113 SRC="FIGDIR/small/631360v1_ufig1.gif" ALT="Figure 1"> View larger version (22K): org.highwire.dtl.DTLVardef@11f6844org.highwire.dtl.DTLVardef@f65124org.highwire.dtl.DTLVardef@1492571org.highwire.dtl.DTLVardef@13931dd_HPS_FORMAT_FIGEXP M_FIG C_FIG HighlightsO_LIHnRNPD mediates oral cancer cell proliferation, migration, invasion, and survival. C_LIO_LIHnRNPD acts as a novel regulator of the PI3K/AKT/mTOR axis by destabilization of PTEN. C_LIO_LINF-{kappa}B/RelA downregulated on knockout of HNRNPD, inhibiting autophagy through downregulating its transcriptional target LCB-II. C_LIO_LIHnRNPD mediates cellular senescence in oral cancer cells. C_LI

cancer biology↗

bioRxiv2024

Guava cv. Allahabad Safeda Chromosome scale assembly and comparative genomics decodes breeder choice marker trait association for pink pulp colour

Deciphering chromosomal length genome assemblies has the potential to unravel an organisms evolutionary relationships and genetic mapping of traits of commercial importance. We assembled guava genome using a hybrid sequencing approach with [~]450x depth Illumina short reads, [~]35x PacBio long reads and Bionano maps to [~]594 MB Scaffold length on 11 pseudo chromosomes ([~]479 MB contig length). Maker pipeline predicted 17,395 genes, 23% greater from earlier draft produced in same cultivar Allahabad Safeda. The genome assembly clarified guava evolutionary history, for example revealing predominance of gene expansion by dispersed duplications, in particular contributing to abundance of monoterpene synthases; and supporting evidence of a whole genome duplication event in guava as in other Myrtaceae. Guava breeders have been aiming to reduce screening time for selecting pink pulp colour progenies using marker-trait associations, but a previous comparative transcriptomics and comparative genomics approach with draft genome assembly to identify the effector gene associated with pink pulp was unsuccessful. Here, genome re-sequencing with Illumina short reads at [~]25x depth of 20 pink fleshed and/or non-coloured guava cultivars and comprehensive analysis for genes in the carotenoid biosynthesis pathway identified structural variations in Phytoene Synthase 2. Further, ddRAD based association mapping in core-collection of 82 coloured and non-coloured genotypes from Indian sub-continent found strong association with the same causal gene. Subsequently, we developed PCR based Indel/SSR breeder friendly marker that can readily be scored in routine agarose gels and empowers accurate selection for seedlings that will produce fruits with pink pulp.

plant biology↗

bioRxiv2024

Testing a Method for Quantifying Microplastic Content in Mytilus californianus

With the proliferation of microplastics within the environment, there has been an increased need to quantify them within organic tissues of aquatic species. Several studies have examined the potential role of mussels in filtering these contaminants, prompting interest in characterizing uptake and retention of microplastics in these species. There are a variety of methods to isolate and quantify microplastics, including density separation, enzymatic digestion, and filtration. The applicability of a combined digestion and ultrasonic bath methodology in isolating microplastics from the California mussel (Mytilus californianus Conrad, 1837) was investigated. Microplastics were added to 20-gallon tanks at 0 g (control) 8 mg. Mussel tissues from each tank were digested using potassium hydroxide solutions. These digestions were subjected to vacuum filtration and a sonication bath, then treated with a mixture of sodium hypochlorite and nitric acid. It was found that this method successfully removed organic tissue from the mussels while leaving dyed microplastics intact. The latter were quantified using fluorescent microscopy. This proof-of-principle experiment indicates that this technique can be applied to further studies on environmental impact and retention of microplastics in mussels and may also have utility for other shellfish.

zoology↗

bioRxiv2023

Uncovering Invariant Representations in Functional Neuroimaging with Deep Metric Learning

With the increasing ability to record neuroimaging with higher spatial and temporal resolution, there is a growing need for methods that reduce these high-dimensional representations into latent low-dimensional structures that are discriminative and/or predictive of behavior, disease, or in general experimental context. We propose a metric learning framework to extract meaningful latent structures from high-dimensional fMRI data. This method learns the latent embeddings that reduce the intra-group variability while maximizing the inter-group variability. In addition, our method leverages advances in few-shot learning approaches to adapt to small sample-size fMRI datasets, allowing one to learn the latent structure from just a few samples per context. We evaluate our work on two publicly available fMRI datasets and report superior results compared to popular alternative approaches such as Principal Component Analysis (84.7% vs. 60%; 21.8% vs. 8.3%). We provide the Python code as open-source at Github.

bioinformatics↗

bioRxiv2023

Artificial Intelligence uncovers Evolutionarily Conserved Intracellular Allosteric Modulators of GPCR-Gα interface.

Endogenous intracellular allosteric modulators of GPCRs remain largely unexplored, with limited binding and phenotype data available. This gap arises from the lack of robust computational methods for unbiased cavity identification, cavity-specific ligand design, synthesis, and validation across GPCR topology. Here, we developed Gcoupler, an AI-driven generalized computational toolkit that leverages an integrative approach combining de novo ligand design, statistical methods, Graph Neural Networks, and Bioactivity-based ligand prioritization for rationally predicting high-affinity ligands. Using Gcoupler, we interrogated intracellular metabolites that target and regulate the GPCR-G interface (Ste2p-Gpa1p), affecting pheromone-induced programmed cell death in yeast. Our computational analysis, complemented by experimental validations, including genetic screening, multi-omics, site-directed mutagenesis, biochemical assays, and physiological readouts, identified endogenous hydrophobic metabolites, notably sterols, as direct intracellular allosteric modulators of Ste2p. Molecular simulations coupled with biochemical signaling assessment in site-directed Ste2p mutants further confirmed that metabolites binding to GPCR-G obstruct downstream signaling, possibly via a cohesive effect. Finally, by utilizing isoproterenol-induced, GPCR-mediated human and neonatal rat cardiac hypertrophy models, we observed that elevated metabolite levels attenuate hypertrophic response, reinforcing the evolutionary relevance of this mechanism.

bioinformatics↗

bioRxiv2022

Evidence of kinesin motors involved in stable kinetochore assembly during early meiosis

The characteristic bi-lobed organization of the kinetochores observed during mitotic metaphase is a result of separation of the sister kinetochores into two clusters upon their stable end-on attachment to the microtubules emanating from opposite spindle poles. In contrast, during metaphase I of meiosis despite bi-orientation of the homologs, we observe that the kinetochores are linearly dispersed between the two spindle poles indicating that pole-distal and pole-proximal kinetochores are attached laterally and end-on, respectively to the microtubules. Colocalization studies of kinetochores and kinesin motors suggest that budding yeast kinesin 5, Cin8 and Kip1 perhaps localize to the end-on attached kinetochores while kinesin 8, Kip3 resides at all the kinetochores. Unlike mitosis in budding yeast, in meiosis, the outer kinetochores assemble much later after prophase I. From the findings including co-appearance of kinesin 5 and the outer kinetochore protein Ndc80 at the centromeres after prophase I and a reduction in Ndc80 level in Cin8 null mutant, we propose that kinesin motors are required for reassembly and stability of the kinetochores during early meiosis. Thus, this work reports yet another meiosis specific function of kinesin motor.

cell biology↗