Biology subjects

Malladi, V. S.

Publications and source records attributed to Malladi, V. S..

8 recordsLinked to original sources

bioRxiv2026

The GA4GH Workflow Execution Service: A Standard API for Workflow Execution and Management in Genomics

Rapid growth in life science data generation has led to large and increasingly distributed collections of sensitive genomic data. To analyse this data, researchers increasingly rely on computational workflows to enable scalable, portable and reproducible analysis. However, these workflows are executed by specialized workflow engines across heterogeneous compute environments, each with its own requirements. As a result, users often need engine- and compute-environment-specific knowledge to submit, monitor and manage workflow execution, increasing the barriers to interoperability. These challenges are compounded in large-scale projects, where workflow runs must be dispatched to local clusters, cloud platforms or data hosting infrastructures with different underlying technologies. We describe the Global Alliance for Genomics and Health (GA4GH) Workflow Execution Service (WES) API, an approved GA4GH standard first released in 2018, together with its data model and the endpoints that define a unified interface for workflow execution. By standardizing workflow invocation, the WES API allows a single client to submit, monitor and cancel workflow runs across diverse workflow engines and compute backends, such as scheduling clusters or cloud platforms. We surveyed thirteen adoptions of the standard, covering standalone services and platforms in four programming languages. In a two-site demonstration, a WES implementation from Germany and one from Japan executed the same federated allele-frequency analysis with site-specific adjustments and only aggregated results leaving either site, demonstrating that it is possible to run the same analysis across independently built and operated WES services in different countries. From the gaps identified, we propose extensions for community discussion.

bioinformatics↗

bioRxiv2025

X-Linked Cancer-Associated Polypeptide (XCP) from lncRNA1456 Cooperates with PHF8 to Regulate Gene Expression and Cellular Pathways in Breast Cancer

Recent studies have demonstrated that a subset of long "noncoding" RNAs (lncRNAs) produce functional polypeptides and proteins. In this study, we discovered a 132 amino acid protein in human breast cancer cells named XCP (X-linked Cancer-associated Polypeptide), which is encoded by lncRNA1456 (a.k.a. RHOXF1P3), a transcript previously thought to be noncoding. lncRNA1456 is a pancreas- and testis-specific RNA whose gene is located on chromosome X. We found that the expression of lncRNA1456 and XCP are highly upregulated in the luminal A, luminal B, and HER2 molecular subtypes of breast cancer. XCP modulates both estrogen-dependent and estrogen-independent growth of breast cancer cells by regulating cancer pathways, as shown in cell and xenograft models. XCP shares some homology with homeodomain-containing proteins and interacts with the histone demethylase plant homeodomain finger protein 8 (PHF8), which is also encoded by an X-linked gene. Mechanistically, XCP stimulates the histone demethylase activity of PHF8 to regulate gene expression in breast cancer cells. These findings identify XCP as a coregulator of transcription and emphasize the need to interrogate the potential functional roles of open reading frames originating from noncoding RNAs. Statement of SignificangeXCP, a polypeptide encoded by an X-linked lncRNA, regulates gene expression in breast cancer cells. XCP is a chromatin-associated protein that interacts with the histone demethylase PHF8 and modulates its demethylase activity to regulate gene expression. XCP drives cancer-specific phenotypes and serves as a potential biomarker and/or target for therapeutic intervention.

cancer biology↗

bioRxiv2025

sciLaMA: A Single-Cell Representation Learning Framework to Leverage Prior Knowledge from Large Language Models

Single-cell RNA sequencing (scRNA-seq) enables high-resolution exploration of cellular diversity and gene regulation, yet analyzing such data remains challenging due to technical and methodological limitations. Existing task-specific deep generative models like Variational Auto-Encoder (VAE) and its variants struggle to incorporate external biological knowledge, while transformer-based foundational large Language Models (LLMs or large LaMs) face limitations in computational cost and applicability to tabular gene expression data. Here, we introduce sciL-aMA (single-cell interpretable Language Model Adapter), a novel representation learning frame-work that bridges these gaps by integrating static gene embeddings from multimodal LLMs with scRNA-seq tabular data through a paired-VAE architecture. Our approach generates context-aware representations for both cells and genes and outperforms state-of-the-art methods in key single-cell downstream tasks, including batch effect correction, cell clustering, and cell-state-specific gene marker and module identification, while maintaining computational efficiency. sciL-aMA offers a computationally efficient, unified framework for comprehensive single-cell data analysis and biologically interpretable gene module discovery. Source code is available at https://github.com/microsoft/sciLaMA

bioinformatics↗

bioRxiv2023

The ENCODE Uniform Analysis Pipelines

The Encyclopedia of DNA elements (ENCODE) project is a collaborative effort to create a comprehensive catalog of functional elements in the human genome. The current database comprises more than 19000 functional genomics experiments across more than 1000 cell lines and tissues using a wide array of experimental techniques to study the chromatin structure, regulatory and transcriptional landscape of the Homo sapiens and Mus musculus genomes. All experimental data, metadata, and associated computational analyses created by the ENCODE consortium are submitted to the Data Coordination Center (DCC) for validation, tracking, storage, and distribution to community resources and the scientific community. The ENCODE project has engineered and distributed uniform processing pipelines in order to promote data provenance and reproducibility as well as allow interoperability between genomic resources and other consortia. All data files, reference genome versions, software versions, and parameters used by the pipelines are captured and available via the ENCODE Portal. The pipeline code, developed using Docker and Workflow Description Language (WDL; https://openwdl.org/) is publicly available in GitHub, with images available on Dockerhub (https://hub.docker.com), enabling access to a diverse range of biomedical researchers. ENCODE pipelines maintained and used by the DCC can be installed to run on personal computers, local HPC clusters, or in cloud computing environments via Cromwell. Access to the pipelines and data via the cloud allows small labs the ability to use the data or software without access to institutional compute clusters. Standardization of the computational methodologies for analysis and quality control leads to comparable results from different ENCODE collections - a prerequisite for successful integrative analyses. Database URL: https://www.encodeproject.org/

bioinformatics↗

bioRxiv2021

SeqWho: Reliable, rapid determination of sequence file identity using k-mer frequencies

With the vast improvements in sequencing technologies and increased number of protocols, sequencing is finding more applications to answer complex biological problems. Thus, the amount of publicly available sequencing data has tremendously increased in repositories such as SRA, EGA, and ENCODE. With any large online database, there is a critical need to accurately document study metadata, such as the source protocol and organism. In some cases, this metadata may not be systematically verified by the hosting sites and may result in a negative influence on future studies. Here we present SeqWho, a program designed to heuristically assess the quality of sequencing files and reliably classify the organism and protocol type. This is done in an alignment-free algorithm that leverages a Random Forest classifier to learn from native biases in k-mer frequencies and repeat sequence identities between different sequencing technologies and species. Here, we show that our method can accurately and rapidly distinguish between human and mouse, nine different sequencing technologies, and both together, 98.32%, 97.86%, and 96.38% of the time in high confidence calls respectively. This demonstrates that SeqWho is a powerful method for reliably checking the identity of the sequencing files used in any pipeline and illustrates the programs ability to leverage k-mer biases.

bioinformatics↗

bioRxiv2021

Cancer drug discovery as a low rank tensor completion problem

The heterogeneity of cancer necessitates developing a multitude of targeted therapies. We propose the view that cancer drug discovery is a low rank tensor completion problem. We implement this vision by using heterogeneous public data to construct a tensor of drug-target-disease associations. We show the validity of this approach computationally by simulations, and experimentally by testing drug candidates. Specifically, we show that a novel drug candidate, SU11652, controls melanoma tumor growth, including BRAFWT melanoma. Independently, we show that another molecule, TC-E 5008, controls tumor proliferation on ex vivo ER+ human breast cancer. Most importantly, we identify these chemicals with only a few computationally selected experiments as opposed to brute-force screens. The efficiency of our approach enables use of ex vivo human tumor assays as a primary screening tool. We provide a web server, the Cancer Vulnerability Explorer (accessible at https://cavu.biohpc.swmed.edu), to facilitate the use of our methodology.

cancer biology↗

bioRxiv2020

Urethral luminal epithelia are castration-insensitive progenitors of the proximal prostate

Castration-insensitive epithelial progenitors capable of regenerating the prostate are concentrated in the proximal region close to the urethra, but the identification of these cells has been limited to individual cell surface markers. Here, we use single cell RNA sequencing (scRNA-seq) to obtain a cellular anatomy of the mouse prostate and urethra and create a comparative map with the human. These data reveal that previously identified facultative progenitors marked by TROP2, Sca-1, KRT4, and PSCA are actually luminal epithelia of the urethra that extend into the proximal prostate. These mouse urethral cells are the human equivalent of previously identified club and hillock urethral cells. Castration decreases androgen-dependent prostate luminal epithelia as expected, but TROP2+ urethral luminal epithelia survive and expand into the prostate. Benign prostatic hyperplasia (BPH) has long been considered an embryonic reawakening, but the cellular origin of peri-urethral growth is unclear. We use scRNA-seq and flow cytometry to demonstrate an increase in urethral luminal epithelia within glandular nodules from patients with BPH, which are further enriched in patients treated with a 5 alpha reductase inhibitor. These data demonstrate that the putative prostate progenitors enriched by castration in the proximal prostate are an expansion of urethral luminal epithelia and that these cells may play an important role in the etiology of human BPH. Significance StatementThe prostate involutes after castration, but regrows to its original size with androgen replenishment. This observation prompted the search for a castration-insensitive prostate progenitor. Here, Joseph et al. produce a comparative cellular atlas of the prostate and urethra in the mouse vs. human, discovering an equivalent urethral luminal epithelial cell type that extends into the proximal prostatic ducts and expresses previously identified markers of facultative prostate progenitors. Urethral luminal epithelia are established before prostate budding in human and mouse development, and expand after castration in the mouse and after 5 alpha reductase inhibitor treatment in human BPH. These data suggest that luminal epithelia of the urethra are castration-insensitive cells of proximal ducts that may act as progenitors in human BPH.

cell biology↗

bioRxiv2018

A cellular anatomy of the normal adult human prostate and prostatic urethra

A cellular anatomy of normal human organs is essential for solving the cellular origins of disease. We report the first comprehensive cellular atlas of the young adult human prostate and prostatic urethra using an iterative process of single cell RNA sequencing and flow cytometry on ~98,000 cells taken from different anatomical regions. Two previously unrecognized epithelial cell types were identified by KRT13 and SCGB1A1 expression and found to be highly similar to hillock and club cells of the proximal lung. It was demonstrated by immunohistochemistry that prostate club and hillock cells are similarly concentrated in the proximal prostate. We also optimized a new flow cytometry antibody panel to improve cell type-specific purification based on newly established cellular markers. The molecular classification, anatomical distribution, and purification methods for each cell type in the human prostate create a powerful new resource for experimental design in human prostate disease.

cell biology↗