Purvesh Khatri

The paper designs compact RNA sensors for complex functions of multiple inputs using crowdsourced designs and an algorithm, achieving applications like tuberculosis diagnosis.

The paper identifies a novel framework to quantify immune dysregulation in critical illness by analyzing transcriptomic data across multiple cohorts.

This study identifies cytokines that enhance influenza vaccine antibody responses in human organoids and mice, showing their potential as adjuvants to improve vaccine efficacy.

The paper identifies a conserved immune signature linked to infection severity, risk factors, and treatment response using multi-cohort transcriptomic and proteomic data.

scMetaIntegrator provides a meta-analysis approach for paired scRNA-seq data, improving robustness and reducing false positives by accounting for biological replicates and cell variability.

This paper benchmarks HLA genotype prediction from single-cell RNA-seq data against molecular genotyping, showing improved accuracy with composite models and meta-analytic approaches.

This study uses single-cell transcriptomics to identify gene regulatory networks and lineage trajectories in human fetal thymic epithelial cells, revealing interferon response pathways that distinguish thymic epithelium from other anterior foregut-derived organs.

This study identifies age-related lipidomic changes in neural stem cells using mass spectrometry and shows that lipid modifications affect NSC function and activation.

The paper develops a framework to impute histone modification abundances and infer directional associations between them using neural processes, identifying novel interactions and changes post-vaccination.

The paper identifies gene signatures for human monocyte subsets using public transcriptomes, showing they accurately reflect subset levels in disease states and can be used for cytometry-based sorting.

SIMON is an open-source platform that enables biomedical researchers to apply machine learning algorithms to high-dimensional data without requiring advanced programming skills.

The paper develops a 6-mRNA classifier using non-COVID-19 viral infection data to predict COVID-19 severity, achieving high accuracy in independent cohorts.

The paper evaluates how changes in Gene Ontology (GO) and its annotations over time affect the results of GO enrichment analyses in disease studies.

The paper integrates transcriptome and clinical data to identify overlooked disease relationships, such as between myositis and interstitial cystitis.

The paper develops gene-expression-based prognostic models for 30-day mortality in sepsis using community-onset and hospital-acquired sepsis cohorts, showing improved prediction when combined with clinical scores.

The paper identifies gene annotation bias in biomedical resources and highlights its impact on research focus, advocating for data-driven hypothesis generation to mitigate these biases.

This paper introduces immunoStates, a basis matrix that reduces biological and technical biases in cell-mixture deconvolution by incorporating diverse disease states and microarray platforms.