Biology subjects

Nagpal, S.

Publications and source records attributed to Nagpal, S..

14 recordsLinked to original sources

bioRxiv2026

CSF-Seq enables transcriptome-wide profiling of cerebrospinal fluid and identifies prognostic signature of leptomeningeal disease

Leptomeningeal disease (LMD) is a rapidly fatal complication of systemic cancer for which sensitive diagnostic tools and informative biomarkers remain limited. Here, we introduce CSF-Seq, a method for whole-transcriptome sequencing of cell-free RNA (cfRNA) from human cerebrospinal fluid (CSF), designed to enable molecular profiling of LMD and other central nervous system (CNS) conditions. Using a prospectively collected CSF biobank, we analyzed 125 samples spanning multiple pathologies, including breast and lung LMD, glioblastoma, traumatic brain injury, and non-cancer neurological controls. Through optimized RNA extraction, library preparation, and deep sequencing, CSF-Seq generated robust and reproducible transcriptome-wide profiles despite the low abundance and fragmentation of cfRNA in CSF. CSF transcriptomes exhibited disease-specific expression, separating LMD from non-cancer controls and from non-LMD cancers, independent of CSF collection modality. Tumor-associated epithelial transcripts, including CEACAM6 and MUC1, were consistently enriched in LMD samples, whereas immune and CNS-associated transcripts were broadly detected across disease states, consistent with contributions from both tumor and non-tumor sources. Cross-site processing of matched samples demonstrated high concordance, indicating preservation of sample-specific transcriptional signatures across independent workflows. Importantly, we identified a collection method- independent LMD gene expression signature that was significantly associated with overall survival, supporting its potential prognostic relevance. Together, these findings establish CSF-Seq as a technically robust and clinically informative platform for transcriptomic biomarker discovery in CNS metastatic disease, offering a minimally invasive approach for disease characterization, risk stratification, and longitudinal monitoring in patients with LMD.

cancer biology↗

bioRxiv2026

Tracking vaginal microbiome transitions in bacterial vaginosis for cues of antibiotic resilience

BackgroundBacterial vaginosis (BV) is a common and difficult-to-treat vaginal disorder, with significant implications for reproductive health, particularly in low and middle-income countries. Clinical cure based on symptom resolution or Nugent scores often do not correspond to restoration of healthy vaginal microbiome. Factors underlying treatment failure remain poorly defined, warranting the need for understanding post-treatment microbiome dynamics to improve long-term outcomes. ObjectivesTo delineate longitudinal vaginal bacteriome dynamics, integrating microbial composition, transition patterns, and clinical symptoms in a closely followed cohort of women with BV based on treatment outcome. MethodsVaginal swabs from reproductive-age women (18-45 years) were collected and classified as BV-positive ([≥]7) or healthy ([≤]3) using Nugent scoring. BV cases were treated using single-dose secnidazole and followed for three months. Sociodemographic, clinical, and behavioral data were statistically analyzed across groups. Vaginal microbiome composition was assessed using 16S rRNA sequencing, evaluating taxonomic profiles, alpha and beta diversity, differential abundance, and co-occurrence networks. ResultsAntibiotic treatment reduced overall microbial diversity and shifted community composition toward healthy controls, though relapse samples retained higher diversity of BV-associated taxa such as Sneathia, Dialister, and Gardnerella, while no-relapse and control groups showed higher Lactobacillus abundance. Corynebacterium amycolatum appeared protective, while Mycoplasma and Fusobacterium played symptom-specific roles. Microbial network analysis showed denser and more persistent associations in baseline and relapse groups, with Sneathia remaining a central node. ConclusionShort-term symptom resolution in BV does not correspond to full microbial recovery; necessary for long-term remission. Functional traits of resilient taxa like Sneathia, Fannyhessea and Dialister may confer resilience and enable recolonization, undermining long term treatment efficacy.

microbiology↗

bioRxiv2024

LUNAR: A Deep Learning Model to Predict Glioma Recurrence Using Integrated Genomic and Clinical Data

Gliomas account for approximately 25.5% of all primary brain and central nervous system tumors, with a striking 80.8% of these being malignant. The prognosis varies significantly; low-grade gliomas (LGGs) can exhibit 5-year survival rates of up to 80%, while higher-grade gliomas (HGGs) often see rates below 5%. Recurrence is a common challenge, occurring in 52%-62% of LGGs and 90% of HGGs, complicating clinical management and treatment planning. Currently, no widely available models exist for predicting glioma recurrence, which is critical for optimizing patient outcomes. Machine learning (ML) and deep learning (DL) techniques have shown promise in predicting recurrence for various cancers, most using Electronic Health Records (EHR). This study introduces gLioma recUrreNce Attention-based classifieR (LUNAR), a DL-based model to predict early versus late glioma recurrence by integrating clinical, genomic, and mRNA expression data from patients with primary grade II-IV gliomas. The data was obtained from The Cancer Genome Atlas and the Glioma Longitudinal Analysis Consortium (GLASS).

bioinformatics↗

bioRxiv2023

EnsembleSeq: A workflow towards real-time, rapid and simultaneous multi-kingdom amplicon sequencing for holistic and cost-effective microbiome research at scale

BackgroundBacterial communities are often concomitantly present with numerous microorganisms in the human body and other natural environments. Amplicon based microbiome studies have generally paid a skewed attention, that too at a rather shallow genus level resolution, to the highly abundant bacteriome, with interest now forking towards the other microorganisms, particularly fungi. Given the generally sparse abundance of other microbes in the total microbiome, simultaneous sequencing of amplicons targeting multiple microbial kingdoms could be possible even with full multiplexing. Guiding studies are currently needed for performing and monitoring multi-kingdom-amplicon sequencing and data capture at scale. MethodFull length bacterial 16S rRNA gene and entire fungal ITS region amplification was performed for human saliva samples (n=96, including negative and positive controls). Combined amplicon DNA libraries were prepared for nanopore sequencing using a major fraction of 16S molecules and a minor fraction of ITS amplicons. Sequencing was performed in a single run of an R10.4.1 flowcell employing the latest V14 chemistry. An approach for real time monitoring of the species saturation using dynamic rarefaction was designed as a guiding determinant of optimal run time. ResultsReal-time saturation monitoring for both bacterial and fungal species enabled the completion of sequencing within 30 hours, utilizing less than 60% of the total nanopores. [~]5 million HQ taxonomically assigned reads were generated ([~]4.2 million bacterial and 0.7 million fungal), providing a wider (beyond bacteriome) snapshot of human oral microbiota at species level resolution. Among the more than 400 bacterial and 240 fungal species identified in the studied samples, the species of Streptococcus (e.g. S. mitis, S. oralis) and Candida (e.g. C. albicans, C. tropicalis) were observed to be the dominating microbes in the oral cavity, respectively. This conformed well with the previous reports of the human oral microbiota. ConclusionEnsembleseq provides a proof-of-concept towards identification of both fungal and bacterial species simultaneously in a single fully multiplexed nanopore sequencing run in a time and resource effective manner. Details of this workflow are provided to enable large scale application for a holistic species level microbiome study.

microbiology↗

bioRxiv2023

How to Scan DNA using Promiscuous Recognition and No Sliding Clamp: A Model for Pioneer Transcription Factors

DNA scanning proteins slide on the DNA assisted by a clamping interface and uniquely recognize their cognate sequence motif. The transcription factors that control cell fate in eukaryotes must forgo these elements to gain access to both naked DNA and chromatin, so whether or how they scan DNA is unknown. Here we use single-molecule techniques to investigate DNA scanning by the Engrailed homeodomain (enHD) as paradigm of promiscuous recognition and open DNA interaction. We find that enHD scans DNA as fast and extensively as conventional scanners and 10,000,000 fold faster than expected for a continuous promiscuous slide. Our results indicate that such supercharged scanning involves stochastic alternants between local sequence sweeps of [~]85 bp and very rapid deployments to locations [~]500 bp afar. The scanning mechanism of enHD reveals a strategy perfectly suited for the highly complex environments of eukaryotic cells that might be generally used by pioneer transcription factors. TeaserEukaryotic transcription factors can efficiently scan DNA using a rather special mechanism based on promiscuous recognition.

biophysics↗

bioRxiv2023

Optogenetic control of kinesins -1, -2, -3, and dynein reveals their specific roles in vesicular transport

Each cargo in a cell employs a unique set of motor proteins for its transport. Often multiple types of kinesins are bound to the same cargo. It is puzzling why several types of motors are required for robust transport. To dissect the roles of each type of motor, we developed optogenetic inhibitors of kinesin-1, -2, -3 and dynein. This system allows us to control the activity of the endogenous set of motor proteins that are bound to intracellular cargoes. We examined the effect of optogenetic inhibition of kinesins-1, -2, and -3 and dynein on the transport of early endosomes, late endosomes, and lysosomes. While kinesin-1, kinesin-3, and dynein transport vesicles at all stages of endocytosis, kinesin-2 primarily drives late endosomes and lysosomes. In agreement with previous studies, sustained inhibition of either kinesins or dynein results in reduced motility in both directions. However, transient, optogenetic inhibition of kinesin-1 or dynein causes both early and late endosomes to move more processively by relieving competition with opposing motors. In contrast, optogenetic inhibition of kinesin-2 reduces the motility of late endosomes and lysosomes, and inhibition of kinesin-3 reduces the motility of endosomes and lysosomes. These results suggest that the directionality of transport is likely controlled through regulating kinesin-1 and dynein activity. On vesicles transported by several kinesin and dynein motors, motility can be directed by modulating the activity of a single type of motor on the cargo.

biophysics↗

bioRxiv2021

(Machine) Learning the mutation signatures of SARS-CoV-2: a primer for predictive prognosis

MotivationContinuous emergence of new variants through appearance, accumulation and disappearance of mutations in viruses is a hallmark of many viral diseases. SARS-CoV-2 and its variants have particularly exerted tremendous pressure on global healthcare system owing to their life threatening and debilitating implications. The sheer plurality of the variants and huge scale of genome sequence data available for Covid19 have added to the challenges of traceability of mutations of concern. The latter however provides an opportunity to utilize SARS-CoV-2 genomes and the mutations therein as big data records to comprehensively classify the variants through the (machine) learning of mutation patterns. The unprecedented sequencing effort and tracing of disease outcomes provide an excellent ground for identifying important mutations by developing machine learnt models or severity classifiers using mutation profile of SARS-CoV-2. This is expected to provide a significant impetus to the efforts towards not only identifying the mutations of concern but also exploring the potential of mutation driven predictive prognosis of SARS-CoV-2. ResultsWe describe how a graduated approach of building various severity specific machine learning classifiers, using only the mutation corpus of SARS-CoV-2 genomes, can potentially lead to the identification of important mutations and guide potential prognosis of infection. We demonstrate the applicability of model derived important mutations and use of Shapley values in order to identify the significant mutations of concern as well as for developing sparse models of outcome classification. A total of 77,284 outcome traced SARS-CoV-2 genomes were employed in this study which represented a total corpus of 30346 unique nucleotide mutations and 18647 amino acid mutations. Machine learning models pertaining to graduated classifiers of target outcomes namely Asymptomatic, Mild, Symptomatic/Moderate, Severe and Fatal were built considering the TRIPOD guidelines for predictive prognosis. Shapley values for model linked important mutations were employed to select significant mutations leading to identification of less than 20 outcome driving mutations from each classifier. We additionally describe the significance of adopting a temporal modeling approach to benchmark the predictive prognosis linked with continuously evolving pathogens. A chronologically distinct sampling is important in evaluating the performance of models trained on past data in accurately classifying prognosis linked with genomes of future (observed with new mutations). We conclude that while machine learning approach can play a vital role in identifying relevant mutations, caution should be exercised in using the mutation signatures for predictive prognosis in cases where new mutations have accumulated along with the previously observed mutations of concern. Contactsharmila.mande@tcs.com Supplementary informationSupplementary data are enclosed.

bioinformatics↗

bioRxiv2021

Genomic Surveillance of COVID-19 Variants with Language Models and Machine Learning

The global efforts to control COVID-19 are threatened by the rapid emergence of novel SARS-CoV-2 variants that may display undesirable characteristics such as immune escape, increased transmissibility or pathogenicity. Early prediction for emergence of new strains with these features is critical for pandemic preparedness. We present Strainflow, a supervised and causally predictive model using unsupervised latent space features of SARS-CoV-2 genome sequences. Strainflow was trained and validated on 0.9 million sequences for the period December, 2019 to June, 2021 and the frozen model was prospectively validated from July, 2021 to December, 2021. Strainflow captured the rise in cases two months ahead of the Delta and Omicron surges in most countries including the prediction of a surge in India as early as beginning of November, 2021. Entropy analysis of Strainflow unsupervised embeddings clearly reveals the explore-exploit cycles in genomic feature-space, thus adding interpretability to the deep learning based model. We also conducted codon-level analysis of our model for interpretability and biological validity of our unsupervised features. Strainflow application is openly available as an interactive web-application for prospective genomic surveillance of COVID-19 across the globe.

bioinformatics↗

bioRxiv2021

Stratification of Risk of Progression to Colectomy in Ulcerative Colitis using Measured and Predicted Gene Expression

An important goal of clinical genomics is to be able to estimate the risk of adverse disease outcomes. Between 5% and 10% of ulcerative colitis (UC) patients require colectomy within five years of diagnosis, but polygenic risk scores (PRS) utilizing findings from GWAS are unable to provide meaningful prediction of this adverse status. By contrast, in Crohns disease, gene expression profiling of GWAS-significant genes does provide some stratification of risk of progression to complicated disease in the form of a Transcriptional Risk Score (TRS). Here we demonstrate that both measured (TRS) and polygenic predicted gene expression (PPTRS) identify UC patients at 5-fold elevated risk of colectomy with data from the PROTECT clinical trial and UK Biobank population cohort studies, independently replicated in an NIDDK-IBDGC dataset. Prediction of gene expression from relatively small transcriptome datasets can thus be used in conjunction with transcriptome-wide association studies to stratify risk of disease complications.

genetics↗

bioRxiv2020

What if we perceive SARS-CoV-2 genomes as documents? Topic modelling using Latent Dirichlet Allocation to identify mutation signatures and classify SARS-CoV-2 genomes

Topic modeling is frequently employed for discovering structures (or patterns) in a corpus of documents. Its utility in text-mining and document retrieval tasks in various fields of scientific research is rather well known. An unsupervised machine learning approach, Latent Dirichlet Allocation (LDA) has particularly been utilized for identifying latent (or hidden) topics in document collections and for deciphering the words that define one or more topics using a generative statistical model. Here we describe how SARS-CoV-2 genomic mutation profiles can be structured into a Bag of Words to enable identification of signatures (topics) and their probabilistic distribution across various genomes using LDA. Topic models were generated using ~47000 novel corona virus genomes (considered as documents), leading to identification of 16 amino acid mutation signatures and 18 nucleotide mutation signatures (equivalent to topics) in the corpus of chosen genomes through coherence optimization. The document assumption for genomes also helped in identification of contextual nucleotide mutation signatures in the form of conventional N-grams (e.g. bi-grams and tri-grams). We validated the signatures obtained using LDA driven method against the previously reported recurrent mutations and phylogenetic clades for genomes. Additionally, we report the geographical distribution of the identified mutation signatures in SARS-CoV-2 genomes on the global map. Use of the non-phylogenetic albeit classical approaches like topic modeling and other data centric pattern mining algorithms is therefore proposed for supplementing the efforts towards understanding the genomic diversity of the evolving SARS-CoV-2 genomes (and other pathogens/microbes).

bioinformatics↗

bioRxiv2019

Lymphatic metastases have more diverse roots than distant metastases

Both lymphatic and distant metastases arise through cancer cell migration and colonization of ectopic sites. Nonetheless, the two metastasis types are associated with significantly different clinical outcomes, suggesting that distinct biological mechanisms may drive their formation. Here we show fundamental differences in the seeding patterns of lymphatic and distant metastases. Analyzing the reconstructed phylogenies of human colorectal cancers, we find that distant metastases typically are monophyletic, originating from one common ancestor. Lymphatic metastases, in contrast, are almost exclusively polyphyletic and can be seeded from many primary tumor regions. We develop a rigorous mathematical framework for quantifying the phylogenetic diversity of metastases while accounting for differential lesion sampling among patients. Our results indicate that a smaller fraction of primary tumor cells gives rise to distant metastases than lymphatic metastases. Thus, the two metastasis types exhibit profoundly distinct phylogenetic traits, indicating that different evolutionary mechanisms may drive their formation and influence their clinical behavior.

cancer biology↗

bioRxiv2018

TIGAR: An Improved Bayesian Tool for Transcriptomic Data Imputation Enhances Gene Mapping of Complex Traits

The transcriptome-wide association studies (TWAS) that test for association between the study trait and the imputed gene expression levels from cis-acting expression quantitative trait loci (cis-eQTL) genotypes have successfully enhanced the discovery of genetic risk loci for complex traits. By using the gene expression imputation models fitted from reference datasets that have both genetic and transcriptomic data, TWAS facilitates gene-based tests with GWAS data while accounting for the reference transcriptomic data. The existing TWAS tools like PrediXcan and FUSION use parametric imputation models that have limitations for modeling the complex genetic architecture of transcriptomic data. Therefore, we propose an improved Bayesian method that assumes a data-driven nonparametric prior to impute gene expression. Our method is general and flexible and includes both the parametric imputation models used by PrediXcan and FUSION as special cases. Our simulation studies showed that the nonparametric Bayesian model improved both imputation R2 for transcriptomic data and the TWAS power over PrediXcan. In real applications, our nonparametric Bayesian method fitted transcriptomic imputation models for 2X number of genes with 1.7X average regression R2 over PrediXcan, thus improving the power of follow-up TWAS. Hence, the nonparametric Bayesian model is preferred for modeling the complex genetic architecture of transcriptomes and is expected to enhance transcriptome-integrated genetic association studies. We implement our Bayesian approach in a convenient software tool \"TIGAR\" (Transcriptome-Integrated Genetic Association Resource), which imputes transcriptomic data and performs subsequent TWAS using individual-level or summary-level GWAS data.

genetics↗

bioRxiv2018

The Medical Genome Reference Bank: Whole genomes and phenotype of 2,570 healthy elderly

Population health research is increasingly focused on the genetic determinants of healthy ageing, but there is no public resource of whole genome sequences and phenotype data from healthy elderly individuals. Here we describe the Medical Genome Reference Bank (MGRB), comprising whole genome sequence and phenotype of 2,570 elderly Australians depleted for cancer, cardiovascular disease, and dementia. We analysed the MGRB for single-nucleotide, indel and structural variation in the nuclear and mitochondrial genomes. Individuals in the MGRB had fewer disease-associated common and rare germline variants, relative to both cancer cases and the gnomAD and UK BioBank cohorts, consistent with risk depletion. Pervasive age-related somatic changes were correlated with grip strength in men, suggesting blood-derived whole genomes may also provide a biologic measure of age-related functional deterioration. The MGRB provides a broadly applicable reference cohort for clinical genetics and genomic association studies, and for understanding the genetics of healthy ageing. This research has been conducted using the UK Biobank Resource under Application Number 17984.

genomics↗

bioRxiv2017

Recurrently Mutated Genes Differ between Leptomeningeal and Solid Lung Cancer Brain Metastases

PurposeBrain metastases from non-small cell lung cancer (NSCLC) engraft and grow either within the brain (solid) or diffusely on its surface (leptomeningeal disease; LMD). Routine clinical diagnostics have low sensitivity and provide no information about the underlying mutations. A recurrent mutation analysis of LMD and a comparison between solid and LMD NSCLC brain metastases have yet to be explored.\n\nExperimental DesignWe performed whole-exome sequencing (WES) on eight cerebrospinal fluid (CSF) specimens from NSCLC LMD patients. We compared our LMD sequencing data with a published data set of 26 NSCLC solid brain metastases to determine the relative mutation frequency. We then performed a retrospective chart review of an additional set of 44 NSCLC LMD patients to further evaluate LMD mutations and clinical prognosis.\n\nResultsSix (75%) LMD cases had mutations in EGFR, while none had KRAS mutations. Retrospective chart review revealed only 4 LMD cases (7.7%) with KRAS mutations, but 33 cases (63.5%) with EGFR mutations. TP53 was mutated in 4/8 LMD (50%) cases and 13/26 of solid metastasis (50%). The median interval for developing LMD from NSCLC was shorter in EGFR-mutant (16.3 mo) than wild-type (23.9 mo) patients (p = 0.017).\n\nConclusionsEGFR and TP53 mutations were frequent in LMD exomes (combined frequency 87.5%), suggesting that PCR-based mutation detection assays towards these two genes could be a useful complement to current diagnostics. Correlations of EGFR in LMD and KRAS in solid metastases suggest molecular distinctions or systemic treatment pressure underpinning differences in growth patterns within the brain.\n\nTranslational RelevanceLeptomeningeal disease is a diffuse, malignant, and incurable metastatic brain tumor that accounts for 5-10% of brain metastases. Patients with LMD do not undergo biopsy and their overall prognosis is poor (median survival 3 to 27 months), making it difficult to collect sufficient samples for recurrent mutation analysis. Standard diagnostic procedures (MRI and cytology) for LMD provide no genetic information. To understand the mutation landscape of LMD, we performed whole-exome sequencing on eight lung-derived LMD cases. We showed that mutations in EGFR occurred more frequently in LMD than solid brain metastases, but KRAS mutations were not present in LMD. Further, mutations in recurrent genes such as EGFR and TP53 could be reliably detected in CSF via droplet digital PCR. Targeted analysis of recurrent mutations thus presents a useful complement to the existing diagnostic toolkit, and differences in mutations between LMD and solid brain metastases suggest distinct molecular mechanisms for growth.

genomics↗