Sricharan Reddy Varra
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Sricharan Reddy Varra has a total of 58 co-authors. Here are the top co-authors by number of shared publications:
Alex Kong
Department of Pathology, Stanford University School of Medicine, Stanford, USA
3 shared publicationsCandace C. Liu
Department of Pathology, Stanford University School of Medicine, Stanford, USA
3 shared publications- Michael Angelo
Department of Pathology, Stanford University School of Medicine, Palo alto, CA, USA
3 shared publications - Noah F. Greenwald
University of California, San Francisco
3 shared publications - Sean C. Bendall
Department of Pathology, Stanford University School of Medicine, Stanford, USA
3 shared publications Cameron Sowers
Department of Pathology, Stanford University School of Medicine, Stanford, USA
2 shared publications
Latest Publications
- Open Access
- 2 June 2026
- Published by American Association for Cancer Research (AACR) in Cancer Research
- Vol. 86 (14) , OF1-OF15
CellFuse Enables Multimodal Integration of Single-Cell and Spatial Proteomics Data for Systems-Level Analysis in Cancer
Analysis of tumors using single-cell and spatial modalities is critical to advance our understanding of cancer. The growth of technologies that enable these studies provides an increasing number of single cell datasets. Integrating such data across studies will increase the impact of individual studies and speed cancer research. Most existing integration approaches are tailored to transcriptomic data and assume large sets of shared features, an assumption that fails for lower dimensional proteomic measurements. Here, we developed CellFuse, a deep learning–based integration framework that unifies antibody-based proteomic datasets, including high-dimensional cytometry, cellular indexing of transcriptomes and epitopes by sequencing, and spatial proteomics data. Leveraging supervised contrastive learning, CellFuse learned a shared embedding space that enabled accurate cross-modality cell-type prediction and robust label transfer across tumor samples and experimental conditions. Applied to datasets spanning peripheral blood, bone marrow, and lymphoma, CellFuse consistently outperformed existing approaches in recovering clinically relevant populations, including rare malignant and immune subsets. In solid tumors, it reconstructed spatially resolved microenvironments, capturing interactions between malignant, stromal, and immune cells that correlated with treatment response. By enabling scalable, modality-agnostic integration, CellFuse provides a powerful tool to uncover prognostic cell states and delineate the architecture of the tumor–immune ecosystem with translational relevance, driving cancer discoveries. Significance: CellFuse is a supervised contrastive learning framework enabling accurate, scalable integration of single-cell proteomic and transcriptomic data, overcoming sparse marker overlap and differing distributions to improve analysis of single-cell data in cancer.
- 2 June 2026
- Published by American Association for Cancer Research (AACR) in Cancer Research
- Vol. 86 (14) , OF1-OF15
Open AccessCellFuse Enables Multimodal Integration of Single-Cell and Spatial Proteomics Data for Systems-Level Analysis in Cancer
Analysis of tumors using single-cell and spatial modalities is critical to advance our understanding of cancer. The growth of technologies that enable these studies provides an increasing number of single cell datasets. Integrating such data across studies will increase the impact of individual studies and speed cancer research. Most existing integration approaches are tailored to transcriptomic data and assume large sets of shared features, an assumption that fails for lower dimensional proteomic measurements. Here, we developed CellFuse, a deep learning–based integration framework that unifies antibody-based proteomic datasets, including high-dimensional cytometry, cellular indexing of transcriptomes and epitopes by sequencing, and spatial proteomics data. Leveraging supervised contrastive learning, CellFuse learned a shared embedding space that enabled accurate cross-modality cell-type prediction and robust label transfer across tumor samples and experimental conditions. Applied to datasets spanning peripheral blood, bone marrow, and lymphoma, CellFuse consistently outperformed existing approaches in recovering clinically relevant populations, including rare malignant and immune subsets. In solid tumors, it reconstructed spatially resolved microenvironments, capturing interactions between malignant, stromal, and immune cells that correlated with treatment response. By enabling scalable, modality-agnostic integration, CellFuse provides a powerful tool to uncover prognostic cell states and delineate the architecture of the tumor–immune ecosystem with translational relevance, driving cancer discoveries. Significance: CellFuse is a supervised contrastive learning framework enabling accurate, scalable integration of single-cell proteomic and transcriptomic data, overcoming sparse marker overlap and differing distributions to improve analysis of single-cell data in cancer.
- Top Cited Papers
- 18 February 2026
- Published by Springer Nature in Nature Cancer
- Vol. 7 (3) , 435-450
Temporal and spatial composition of the tumor microenvironment predicts response to immune checkpoint inhibition in metastatic TNBC
Immune checkpoint inhibition (ICI) benefits only a subset of patients with metastatic triple-negative breast cancer and determinants of response remain unclear. We assembled a longitudinal cohort of 103 female patients from the phase 2 TONIC trial, with samples spanning primary tumors, pretreatment metastases and on-treatment metastases during nivolumab therapy. We profiled 37 proteins in 270 tumors using highly multiplexed imaging and developed SpaceCat, an open-source pipeline that extracts more than 800 imaging features per sample, including cell density, diversity, spatial interactions and functional marker expression. Metastatic but not primary tumors contained features predictive of outcome. Spatial metrics such as immune diversity and T cell infiltration at tumor borders were most informative, while ratios of T cells to cancer cells and PDL1 on myeloid cells were also associated with response. Multivariate models stratified patients with the highest performance on treatment (area under the curve = 0.90). Bulk RNA-seq confirmed the predictive value of on-treatment samples. These findings highlight the value of longitudinal profiling to resolve evolving tumor microenvironment dynamics driving ICI response.
- 18 February 2026
- Published by Springer Nature in Nature Cancer
- Vol. 7 (3) , 435-450
Top Cited PapersTemporal and spatial composition of the tumor microenvironment predicts response to immune checkpoint inhibition in metastatic TNBC
Immune checkpoint inhibition (ICI) benefits only a subset of patients with metastatic triple-negative breast cancer and determinants of response remain unclear. We assembled a longitudinal cohort of 103 female patients from the phase 2 TONIC trial, with samples spanning primary tumors, pretreatment metastases and on-treatment metastases during nivolumab therapy. We profiled 37 proteins in 270 tumors using highly multiplexed imaging and developed SpaceCat, an open-source pipeline that extracts more than 800 imaging features per sample, including cell density, diversity, spatial interactions and functional marker expression. Metastatic but not primary tumors contained features predictive of outcome. Spatial metrics such as immune diversity and T cell infiltration at tumor borders were most informative, while ratios of T cells to cancer cells and PDL1 on myeloid cells were also associated with response. Multivariate models stratified patients with the highest performance on treatment (area under the curve = 0.90). Bulk RNA-seq confirmed the predictive value of on-treatment samples. These findings highlight the value of longitudinal profiling to resolve evolving tumor microenvironment dynamics driving ICI response.
- 8 October 2025
- Published by Springer Nature in Nature Methods
- Vol. 22 (10) , 2161-2170
Automated classification of cellular expression in multiplexed imaging data with Nimbus
Multiplexed imaging offers a powerful approach to characterize the spatial topography of tissues in both health and disease. To analyze such data, the specific combination of markers that are present in each cell must be enumerated to enable accurate phenotyping, a process that often relies on unsupervised clustering. We constructed the Pan-Multiplex (Pan-M) dataset containing 197 million distinct annotations of marker expression across 15 different cell types. We used Pan-M to create Nimbus, a deep learning model to predict marker positivity from multiplexed image data. Nimbus is a pretrained model that uses the underlying images to classify marker expression of individual cells as positive or negative across distinct cell types, from different tissues, acquired using different microscope platforms, without requiring any retraining. We demonstrate that Nimbus predictions capture the underlying staining patterns of the full diversity of markers present in Pan-M, and that Nimbus matches or exceeds the accuracy of previous approaches that must be retrained on each dataset. We then show how Nimbus predictions can be integrated with downstream clustering algorithms to robustly identify cell subtypes in image data. We have open-sourced Nimbus and Pan-M to enable community use at https://github.com/angelolab/Nimbus-Inference.
- 8 October 2025
- Published by Springer Nature in Nature Methods
- Vol. 22 (10) , 2161-2170
Automated classification of cellular expression in multiplexed imaging data with Nimbus
Multiplexed imaging offers a powerful approach to characterize the spatial topography of tissues in both health and disease. To analyze such data, the specific combination of markers that are present in each cell must be enumerated to enable accurate phenotyping, a process that often relies on unsupervised clustering. We constructed the Pan-Multiplex (Pan-M) dataset containing 197 million distinct annotations of marker expression across 15 different cell types. We used Pan-M to create Nimbus, a deep learning model to predict marker positivity from multiplexed image data. Nimbus is a pretrained model that uses the underlying images to classify marker expression of individual cells as positive or negative across distinct cell types, from different tissues, acquired using different microscope platforms, without requiring any retraining. We demonstrate that Nimbus predictions capture the underlying staining patterns of the full diversity of markers present in Pan-M, and that Nimbus matches or exceeds the accuracy of previous approaches that must be retrained on each dataset. We then show how Nimbus predictions can be integrated with downstream clustering algorithms to robustly identify cell subtypes in image data. We have open-sourced Nimbus and Pan-M to enable community use at https://github.com/angelolab/Nimbus-Inference.
- Top Cited PapersOpen Access
- 1 August 2023
- Published by Springer Nature in Nature Communications
- Vol. 14 (1) , 1-16
Robust phenotyping of highly multiplexed tissue imaging data using pixel-level clustering
While technologies for multiplexed imaging have provided an unprecedented understanding of tissue composition in health and disease, interpreting this data remains a significant computational challenge. To understand the spatial organization of tissue and how it relates to disease processes, imaging studies typically focus on cell-level phenotypes. However, images can capture biologically important objects that are outside of cells, such as the extracellular matrix. Here, we describe a pipeline, Pixie, that achieves robust and quantitative annotation of pixel-level features using unsupervised clustering and show its application across a variety of biological contexts and multiplexed imaging platforms. Furthermore, current cell phenotyping strategies that rely on unsupervised clustering can be labor intensive and require large amounts of manual cluster adjustments. We demonstrate how pixel clusters that lie within cells can be used to improve cell annotations. We comprehensively evaluate pre-processing steps and parameter choices to optimize clustering performance and quantify the reproducibility of our method. Importantly, Pixie is open source and easily customizable through a user-friendly interface.
- 1 August 2023
- Published by Springer Nature in Nature Communications
- Vol. 14 (1) , 1-16
Top Cited PapersOpen AccessRobust phenotyping of highly multiplexed tissue imaging data using pixel-level clustering
While technologies for multiplexed imaging have provided an unprecedented understanding of tissue composition in health and disease, interpreting this data remains a significant computational challenge. To understand the spatial organization of tissue and how it relates to disease processes, imaging studies typically focus on cell-level phenotypes. However, images can capture biologically important objects that are outside of cells, such as the extracellular matrix. Here, we describe a pipeline, Pixie, that achieves robust and quantitative annotation of pixel-level features using unsupervised clustering and show its application across a variety of biological contexts and multiplexed imaging platforms. Furthermore, current cell phenotyping strategies that rely on unsupervised clustering can be labor intensive and require large amounts of manual cluster adjustments. We demonstrate how pixel clusters that lie within cells can be used to improve cell annotations. We comprehensively evaluate pre-processing steps and parameter choices to optimize clustering performance and quantify the reproducibility of our method. Importantly, Pixie is open source and easily customizable through a user-friendly interface.