Biology subjects

Srinivasan, A.

Publications and source records attributed to Srinivasan, A..

35 records · Page 2Linked to original sources

bioRxiv2024

Putting computational models of immunity to the test - an invited challenge to predict B. pertussis vaccination outcomes

Systems vaccinology studies have been used to build computational models that predict individual vaccine responses and identify the factors contributing to differences in outcome. Comparing such models is challenging due to variability in study designs. To address this, we established a community resource to compare models predicting B. pertussis booster responses and generate experimental data for the explicit purpose of model evaluation. We here describe our second computational prediction challenge using this resource, where we benchmarked 49 algorithms from 53 scientists. We found that the most successful models stood out in their handling of nonlinearities, reducing large feature sets to representative subsets, and advanced data preprocessing. In contrast, we found that models adopted from literature that were developed to predict vaccine antibody responses in other settings performed poorly, reinforcing the need for purpose-built models. Overall, this demonstrates the value of purpose-generated datasets for rigorous and open model evaluations to identify features that improve the reliability and applicability of computational models in vaccine response prediction.

immunology↗

bioRxiv2024

Identifying Intermolecular Interactions in Single-Molecule Localization Microscopy

Intermolecular interactions underlie all cellular functions, yet visualizing these interactions at the single-molecule level remains challenging. Single-molecule localization microscopy (SMLM) offers a potential solution. Given a nanoscale map of two putative interaction partners, it should be possible to assign molecules either to the class of coupled pairs or to the class of non-coupled bystanders. Here, we developed a probabilistic algorithm that allows accurate determination of both the absolute number and the proportion of molecules that form coupled pairs. The algorithm calculates interaction probabilities for all possible pairs of localized molecules, selects the most likely interaction set, and corrects for any spurious colocalizations. Benchmarking this approach across a set of simulated molecular localization maps with varying densities (up to [~] 50 molecules {micro}m-2) and localization precisions (5 to 50 nm) showed typical errors in the identification of correct pairs of only a few percent. At molecular densities of [~] 5-10 molecules {micro}m-2 and localization precisions of 20-30 nm, which are typical parameters for SMLM imaging, the recall was [~] 90%. The algorithm was effective at differentiating between non-interacting and coupled molecules both in simulations and experiments. Finally, it correctly inferred the number of coupled pairs over time in a simulated reaction-diffusion system, enabling determination of the underlying rate constants. The proposed approach promises to enable direct visualization and quantification of intermolecular interactions using SMLM.

biophysics↗

bioRxiv2024

Stochastic Voronoi Tessellations as Models for Cellular Neighborhoods in Simple Multicellular Organisms

Recent work on distinct multicellular organisms has revealed a hitherto unknown type of biological noise; rather than a regular arrangement, cellular neighborhood volumes, obtained by Voronoi tessellations of the cell locations, are broadly distributed and consistent with gamma distributions. We propose an explanation for those observations in the case of the alga Volvox, whose somatic cells are embedded in an extracellular matrix (ECM) they export. Both a solvable one-dimensional model of ECM growth derived from bursty transcriptional activity and a two-dimensional "Voronoi liquid" model are shown to provide one-parameter families that smoothly interpolate between the empirically-observed near-maximum-entropy gamma distributions and the crystalline limit of Gaussian distributions governed by the central limit theorem. These results highlight a universal consequence of intrinsic biological noise on the architecture of certain tissues.

biophysics↗

bioRxiv2023

Sustainable production of 2,3,5,6-Tetramethylpyrazine at high titer in engineered Corynebacterium glutamicum using statistical design of experiments

BackgroundThe industrial amino acid production workhorse, Corynebacterium glutamicum naturally produces low levels of 2,3,5,6-tetramethylpyrazine (TMP), a valuable flavor, fragrance and commodity chemical. We have previously reported serendipitous production of TMP in C. glutamicum strains. In the present study, we demonstrate TMP production in C. glutamicum type strain ATCC13032 via the expression of a heterologous TMP pathway in a defined medium followed by statistical design of experiments to understand the effect of the media composition on TMP production. ResultsThe C. glutamicum strain engineered to overexpress acetolactate synthase and alpha-acetolactate decarboxylase from Lactococcus lactis produced [~]0.8 g/L TMP in CGXII minimal medium supplemented with 40 g/L glucose in 24-deep well plates. This engineered strain also demonstrated growth and TMP production when the minimal medium was supplemented with up to 40% (v/v) hydrolysates derived from ionic liquid pretreated sorghum biomass. A screen for improvements in media composition on TMP titer was conducted using fractional factorial design that identified glucose and urea as significant components affecting TMP production. These two components were further optimized using response surface methodology. In the optimized CGXII medium, the engineered strain could produce up to 3.56 g/L TMP (4-fold enhancement in titers and 2-fold enhancement in yield, mol/mol) from 80 g/L glucose and 11.9 g/L urea in shake flask batch cultivation. ConclusionsWe engineered the industrially relevant host, C. glutamicum for targeted production of TMP by heterologous expression of pathway proteins. We demonstrated the capability of the engineered strain for growth and TMP production utilizing real world carbon streams such as hydrolysates. We further identified glucose and urea as the key minimal media components significantly affecting TMP production using statistical media optimization.

synthetic biology↗

bioRxiv2023

Generating Novel Leads for Drug Discovery using LLMs with Logical Feedback

Large Language Models (LLMs) can be used as repositories of biological and chemical information to generate pharmacological lead compounds. However, for LLMs to focus on specific drug targets typically require experimentation with progressively more refined prompts. Results thus become dependent not just on what is known about the target, but also on what is known about the prompt-engineering. In this paper, we separate the prompt into domain-constraints that can be written in a standard logical form, and a simple text-based query. We investigate whether LLMs can be guided, not by refining prompts manually, but by refining the the logical component automatically, keeping the query unchanged. We describe an iterative procedure LMLF ("Language Models with Logical Feedback") in which the constraints are progressively refined using a logical notion of generalisation. On any iteration, newly generated instances are verified against the constraint, providing "logical-feedback" for the next iterations refinement of the constraints. We evaluate LMLF using two well-known targets (inhibition of the Janus Kinase 2; and Dopamine Receptor D2); and two different LLMs (GPT-3 and PaLM). We show that LMLF, starting with the same logical constraints and query text, can guide both LLMs to generate potential leads. We find: (a) Binding affinities of LMLF-generated molecules are skewed towards higher binding affinities than those from existing baselines; LMLF results in generating molecules that are skewed towards higher binding affinities than without logical feedback; (c) Assessment by a computational chemist suggests that LMLF generated compounds may be novel inhibitors. These findings suggest that LLMs with logical feedback may provide a mechanism for generating new leads without requiring the domain-specialist to acquire sophisticated skills in prompt-engineering.

bioinformatics↗

bioRxiv2023

A dose-response based model for statistical analysis of chemical genetic interactions in CRISPRi libraries

An important application of CRISPR interference (CRISPRi) technology is for identifying chemical-genetic interactions (CGIs). Discovery of genes that interact with exposure to antibiotics can yield insights to drug targets and mechanisms of action or resistance. The objective is to identify CRISPRi mutants whose relative abundance is suppressed (or enriched) in the presence of a drug when the target protein is depleted, reflecting synergistic behavior. Different sgRNAs for a given target can induce a wide range of protein depletion and differential effects on growth rate. The effect of sgRNA strength can be partially predicted based on sequence features. However, the actual growth phenotype depends on the sensitivity of cells to depletion of the target protein. For essential genes, sgRNA efficiency can be empirically measured by quantifying effects on growth rate. We observe that the most efficient sgRNAs are not always optimal for detecting synergies with drugs. sgRNA efficiency interacts in a non-linear way with drug sensitivity, producing an effect where the concentration-dependence is maximized for sgRNAs of intermediate strength (and less so for sgRNAs that induce too much or too little target depletion). To capture this interaction, we propose a novel statistical method called CRISPRi-DR (for Dose-Response model) that incorporates both sgRNA efficiencies and drug concentrations in a modified dose-response equation. We use CRISPRi-DR to re-analyze data from a recent CGI experiment in Mycobacterium tuberculosis to identify genes that interact with antibiotics. This approach can be generalized to non-CGI datasets, which we show via an CRISPRi dataset for E. coli growth on different carbon sources. The performance is competitive with the best of several related analytical methods. However, for noisier datasets, some of these methods generate far more significant interactions, likely including many false positives, whereas CRISPRi-DR maintains higher precision, which we observed in both empirical and simulated data. Author SummaryCRISPRi technology is revolutionizing research in various areas of the life sciences, including microbiology, affording the ability to partially deplete the expression of target proteins in a specific and controlled way. Among the applications of CRISPRi, it can be used to construct large (even genome-wide) libraries of knock-down mutants for profiling antibacterial inhibitors and identifying chemical-genetic interactions (CGIs), which can yield insights on drug targets and mechanisms of action and resistance. The data generated by these experiments (i.e., sgRNA counts from high throughput sequencing) is voluminous and subject to various sources of noise. The goal of statistical analysis of such data is to identify significant CGIs, which are genes whose depletion sensitizes cells to an inhibitor. In this paper, we show how to incorporate both sgRNA efficiency and drug concentration simultaneously in a model (CRISPRi-DR) based on an extension of the classic dose-response (Hill) equation in enzymology. This model has advantages over other analytical methods for CRISPRi, which we show using empirical and simulated data.

bioinformatics↗

bioRxiv2023

Development of a mobile, high-throughput, and low-cost image-based plant growth phenotyping system

Nondestructive plant phenotyping is fundamental for unraveling molecular processes underlying plant development and response to the environment. While the emergence of high-through phenotyping facilities can further our understanding of plant development and stress responses, their high costs significantly hinder scientific progress. To democratize high-throughput plant phenotyping, we developed sets of low-cost image- and weight-based devices to monitor plant growth and evapotranspiration. We paired these devices with a suite of computational pipelines for integrated and straightforward data analysis. We validated the suitability of our system for large screens by evaluating a cowpea diversity panel for responses to drought stress. The observed natural variation was subsequently used for Genome-Wide Association Study, where we identified nine genetic loci that putatively contribute to cowpea drought resilience during early vegetative development. We validated the homologs of the identified candidate genes in Arabidopsis using available mutant lines. These results demonstrate the varied applicability of this low-cost phenotyping system. In the future, we foresee these setups facilitating identification of genetic components of growth, plant architecture, and stress tolerance across a wide variety of species.

plant biology↗

bioRxiv2023

Genome-scale and pathway engineering for the sustainable aviation fuel precursor isoprenol production in Pseudomonas putida

Sustainable aviation fuel (SAF) will significantly impact global warming in the aviation sector, and important SAF targets are emerging. Isoprenol is a precursor for a promising SAF compound DMCO (1,4-dimethylcyclooctane), and has been produced in several engineered microorganisms. Recently, Pseudomonas putida has gained interest as a future host for isoprenol bioproduction as it can utilize carbon sources from inexpensive plant biomass. Here, we engineer metabolically versatile host P. putida for isoprenol production. We employ two computational modeling approaches (Bilevel optimization and Constrained Minimal Cut Sets) to predict gene knockout targets and optimize the "IPP-bypass" pathway in P. putida to maximize isoprenol production. Altogether, the highest isoprenol production titer from P. putida was achieved at 3.5 g/L under fed-batch conditions. This combination of computational modeling and strain engineering on P. putida for an advanced biofuels production has vital significance in enabling a bioproduction process that can use renewable carbon streams.

synthetic biology↗

bioRxiv2022

Robust eIF4B levels undermine invasive growth and immune evasion mechanisms in murine triple negative breast cancer models.

Dysregulated protein synthesis is seen in many aggressive cancers, including metastatic breast cancer. However, the specific contributions of certain translation initiation factors to in vivo disease remain undefined. This is particularly true of eIF4B, an RNA-binding protein and cofactor of the RNA helicase eIF4A and associated eIF4F cap-binding complex. While eIF4A, eIF4G, and eIF4E are well-known to contribute to the progression of many cancer types including metastatic breast cancers, the role played by eIF4B in breast cancer remains relatively unclear. We therefore explored how naturally divergent and experimentally modulated eIF4B levels impact tumor growth and progression in well-characterized murine triple negative breast cancer (TNBC) models. Surprisingly, we found that higher eIF4B levels in mouse and human breast cancers were associated with less aggressive phenotypes. shRNA-mediated eIF4B knockdown in TNBC lines failed to markedly alter proliferation and global translation in the cells in vitro and only modestly hindered their growth as primary mammary tumors growth in mice. However, eIF4B knockdown significantly enhanced invasive growth in vitro and exacerbated both tumor burden and mortality relative to nontargeting shRNA controls in a model of metastatic disease. Analysis of eIF4B levels and breast cancer patient survival reinforced a link to better outcomes. Interestingly, low eIF4B expression was also associated with more formidable immune evasion in vitro and in vivo, implicating a novel immunomodulatory role for this factor in the malignant setting that suggests a mode of action beyond its historical role as a co-activator of eIF4A/F. Significance StatementMetastasis is the leading cause of cancer-related mortality. Despite many advances in our understanding of this complex process and the molecular and cellular events involved, mechanisms that allow secondary tumors to arise and persist remain incompletely understood. Uncharacterized metastatic determinants active at the level of translational control may be exploitable as novel therapy targets or biomarkers predicting a tumors potential for spread and recurrence. Here we describe previously unrecognized consequences of dysregulated eIF4B levels in murine breast cancer that shed light on how this translation initiation factor contributes to disease outcomes. Our findings suggest that eIF4B levels direct metastatic risk and immune evasion, and further study should establish its value in personalized treatment decisions and development of future therapies.

cancer biology↗

bioRxiv2022

Hippocampal and Medial Prefrontal Cortical Maps Represent Episodes and Rules in a Common Task Space

Memory helps us adapt to changing circumstances but needs guidance to retrieve relevant episodes. Episodic memory requires the hippocampus, the prefrontal cortex (PFC) guides memory retrieval, but how their representations interact is unclear. Using state-space analysis of neuronal spiking, we found CA1 and PFC activity within and between rats formed similar, low-dimensional, region-specific "shapes" representing different tasks tested in the same maze. Task shapes were organized by behaviorally salient variables including time and maze start and goal locations. PFC predicted CA1 representations when both regions were needed to solve a spatial memory task, but not in a cue approach task that required neither region. Task demands imposed common dimensions on CA1 and PFC maps whose topologies distinguished episodic and rule-related computations.

animal behavior and cognition↗

bioRxiv2022

Larger cerebral cortex is genetically correlated with greater frontal area and dorsal thickness

Human cortical expansion has occurred non-uniformly across the brain. We assessed the genetic architecture of cortical global expansion and regionalization by comparing two sets of genome-wide association studies of 24 cortical regions with and without adjustment for global measures (i.e. total surface area, mean cortical thickness) using a genetically-informed parcellation in 32,488 adults. We found 393 and 756 significant loci with and without adjusting for globals, respectively, among which 8% and 45% loci were associated with more than one region. Results from analyses without adjustment for globals recounted loci associated with global measures. Genetic factors that contribute to total surface area of the cortex particularly expand anterior/frontal regions, whereas those contributing to thicker cortex predominantly increase dorsal/frontal-parietal thickness. Interactome-based analyses revealed significant overlap of global and regional genetic modules, enriched for neurodevelopmental and immune system pathways. Consideration of global measures is important in understanding the genetic variants underlying cortical morphology.

neuroscience↗

bioRxiv2021

Genomics Characterization of an Engineered Corynebacterium glutamicum in Bioreactor Cultivation under Ionic Liquid Stress

Corynebacterium glutamicum is an ideal microbial chassis for the production of valuable bioproducts including amino acids and next-generation biofuels. Here we resequence engineered isopentenol (IP) producing C. glutamicum BRC-JBEI 1.1.2 strain and assess differential transcriptional profiles using RNA sequencing under industrially relevant conditions including scale transition and compare the presence vs. absence of an ionic liquid, cholinium lysinate ([Ch][Lys]). Analysis of the scale transition from shake flask to bioreactor with transcriptomics identified a distinct pattern of metabolic and regulatory responses needed for growth in this industrial format. These differential changes in gene expression corroborate altered accumulation of organic acids and bioproducts, including succinate, acetate, and acetoin that occur when cells are grown in the presence of 50mM [Ch][Lys] in the stirred-tank reactor. This new genome assembly and differential expression analysis of cells grown in a stirred tank bioreactor clarify the cell response of a C. glutamicum strain engineered to produce IP.

systems biology↗

bioRxiv2021

Using Domain-Knowledge to Assist Lead Discovery in Early-Stage Drug Design

We are interested in generating new small molecules which could act as inhibitors of a biological target, when there is limited prior information on target-specific inhibitors. This form of drug-design is assuming increasing importance with the advent of new disease threats for which known chemicals only provide limited information about target inhibition. In this paper, we propose the combined use of deep neural networks and Inductive Logic Programming (ILP) that allows the use of symbolic domain-knowledge (B) to explore the large space of possible molecules. Assuming molecules and their activities to be instances of random variables X and Y, the problem is to draw instances from the conditional distribution of X, given Y, B (DX|Y,B). We decompose this into the constituent parts of obtaining the distributions DX|B and DY|X,B, and describe the design and implementation of models to approximate the distributions. The design consists of generators (to approximate DX|B and DX|Y,B) and a discriminator (to approximate DY|X,B). We investigate our approach using the well-studied problem of inhibitors for the Janus kinase (JAK) class of proteins. We assume first that if no data on inhibitors are available for a target protein (JAK2), but a small numbers of inhibitors are known for homologous proteins (JAK1, JAK3 and TYK2). We show that the inclusion of relational domain-knowledge results in a potentially more effective generator of inhibitors than simple random sampling from the space of molecules or a generator without access to symbolic relations. The results suggest a way of combining symbolic domain-knowledge and deep generative models to constrain the exploration of the chemical space of molecules, when there is limited information on target-inhibitors. We also show how samples from the conditional generator can be used to identify potentially novel target inhibitors.

molecular biology↗

bioRxiv2020

Long-read genome sequencing and assembly of Leptopilina boulardi: a specialist Drosophila parasitoid

Leptopilina boulardi (Hymenoptera: Figitidae) is a specialist parasitoid of Drosophila. The Drosophila-Leptopilina system has emerged as a suitable model for understanding several aspects of host-parasitoid biology. However, a good quality genome of the wasp counterpart was lacking. Here, we report a whole-genome assembly of L. boulardi to bring it in the scope of the applied and fundamental research on Drosophila parasitoids with access to epigenomics and genome editing tools. The 375Mb draft genome has an N50 of 275Kb with 6315 scaffolds >500bp and encompasses >95% complete BUSCOs. Using a combination of ab-initio and RNA-Seq based methods, 25259 protein-coding genes were predicted and 90% (22729) of them could be annotated with at least one function. We demonstrate the quality of the assembled genome by recapitulating the phylogenetic relationship of L. boulardi with other Hymenopterans. The key developmental regulators like Hox genes and sex determination genes are well conserved in L. boulardi, and so is the basic toolkit for epigenetic regulation. The search for epigenetic regulators has also revealed that L. boulardi genome possesses DNMT1 (maintenance DNA methyltransferase), DNMT2 (tRNA methyltransferase) but lacks the de novo DNA methyltransferase (DNMT3). Also, the heterochromatin protein 1 family appears to have expanded as compared to other hymenopterans. The draft genome of L. boulardi (Lb17) will expedite the research on Drosophila parasitoids. This genome resource and early indication of epigenetic aspects in its specialization make it an interesting system to address a variety of questions on host-parasitoid biology.

genomics↗

bioRxiv2019

Compositional knockoff filter for high-dimensional regression analysis of microbiome data

SO_SCPLOWUMMARYC_SCPLOWA critical task in microbiome data analysis is to explore the association between a scalar response of interest and a large number of microbial taxa that are summarized as compositional data at different taxonomic levels. Motivated by fine-mapping of the microbiome, we propose a two-step compositional knockoff filter (CKF) to provide the effective finite-sample false discovery rate (FDR) control in high-dimensional linear log-contrast regression analysis of microbiome compositional data. In the first step, we employ the compositional screening procedure to remove insignificant microbial taxa while retaining the essential sum-to-zero constraint. In the second step, we extend the knockoff filter to identify the significant microbial taxa in the sparse regression model for compositional data. Thereby, a subset of the microbes is selected from the high-dimensional microbial taxa as related to the response using a pre-specified FDR threshold. We study the asymptotic properties of the proposed two-step procedure, including both sure screening and effective false discovery control. We demonstrate the finite-sample properties in simulation studies, which show the gain in the empirical power while controlling the nominal FDR. The potential usefulness of the proposed method is also illustrated with application to an inflammatory bowel disease dataset to identify microbial taxa that influence host gene expressions.

bioinformatics↗

bioRxiv2019

CHOLESTEROL MYTH AND THE IMPERATIVE ROLE OF FUNCTIONAL EVALUATION OF LIPOPROTEINS IN ATHEROSCLEROTIC RISK PREDICTION STRATEGIES

ObjectiveTo arrive at apposite risk prediction strategies in identifying markers that could delineate patients susceptible for atherosclerotic progression from other diabetic patient population thus circumventing undesirable drug exposure or conversely inadequate treatment. Research Design and MethodsThe study was designed to assess and compare the biochemical parameters along with lipoprotein ratios assessed in structural and functional aspects between the probable risk factor groups for CAD(diabetics with no symptoms for CAD) and untreated CAD patients (on admission for CAD) in a total of 593 participants. Anthropometric measurements were made and serum fasting glucose, lipid peroxide levels, lipid and lipoprotein ratio assessments were carried out. In addition to these basic parameters, oxidized LDL (Ox-LDL) and serum paraoxonase (PON) activity were also measured in this study. To ascertain the diagnostic accuracy of the individual markers, receiver operating characteristic curve analysis was carried out. ResultsSerum levels of blood lipids, lipid peroxides were higher in the diabetic and the cardiovascular patients with a concomitant decrease in the levels of HDL. Increased median levels of Ox-LDL and decreased PON activity was observed in the cardiovascular patients both with and without diabetes. ConclusionThis study proposes Ox-LDL/PON assessment would be more functional than the traditional LDL-c/HDL-c to categorize the risk group of patients into prospective atherosclerotic predisposed and non-predisposed population, thus enabling the former to get treated with appropriate medication.

biochemistry↗

bioRxiv2018

Scopoletin Augments DJ1/Nrf2 Signalling and Prevents Protein Aggregation in Parkinson’s disease

Given the role of oxidative stress in PD pathogenesis and off-target side effects of currently available drugs, several natural phytochemicals seems to be promising in the management of PD. Here, we tested the hypothesis that scopoletin, an active principle obtained from Morinda citrifolia (MC), efficiently quenches oxidative stress through DJ-1/Nrf2 signalling and ameliorates rotenone-induced PD. Despite reducing the oxidative stress, administration of MC extract (MCE) has lessened the protein aggregation as evident from decreased levels of nitrotyrosine and -Synuclein. In vitro studies revealed that scopoletin lessened rotenone-induced apoptosis in SH-SY5Y cells through preventing oxidative injury. Particularly, scopoletin markedly up-regulated DJ-1, which then promoted the nuclear translocation of Nrf2 and transactivation of antioxidant genes. Furthermore, we found that scopoletin prevent the nuclear exportation of Nrf2 by reducing the levels of Keap1 and thereby enhancing the neuronal defence system. Overall, our findings suggest that scopoletin acts through DJ-1 mediated Nrf2 signalling to protect the brain from rotenone-induced oxidative stress and PD. Thus, we postulate that scopoletin could be a potential drug to treat PD.

biochemistry↗