Biology subjects

Sorger, P. K.

Publications and source records attributed to Sorger, P. K..

At least 73 records · Page 4Linked to original sources

bioRxiv2021

The spatial landscape of progression and immunoediting in primary melanoma at single cell resolution

Cutaneous melanoma is a highly immunogenic malignancy, surgically curable at early stages, but life- threatening when metastatic. Here we integrate high-plex imaging, 3D high-resolution microscopy, and spatially-resolved micro-region transcriptomics to study immune evasion and immunoediting in primary melanoma. We find that recurrent cellular neighborhoods involving tumor, immune, and stromal cells change significantly along a progression axis involving precursor states, melanoma in situ, and invasive tumor. Hallmarks of immunosuppression are already detectable in precursor regions. When tumors become locally invasive, a consolidated and spatially restricted suppressive environment forms along the tumor-stromal boundary. This environment is established by cytokine gradients that promote expression of MHC-II and IDO1, and by PD1-PDL1 mediated cell contacts involving macrophages, dendritic cells, and T cells. A few millimeters away, cytotoxic T cells synapse with melanoma cells in fields of tumor regression. Thus, invasion and immunoediting can co-exist within a few millimeters of each other in a single specimen. STATEMENT OF SIGNIFICANCEThe reorganization of the tumor ecosystem in primary melanoma is an excellent setting in which to study immunoediting and immune evasion. Guided by classical histopathology, spatial profiling of proteins and mRNA reveals recurrent morphological and molecular features of tumor evolution that involve localized paracrine cytokine signaling and direct cell-cell contact.

cancer biology↗

bioRxiv2021

Fides: Reliable Trust-Region Optimization for Parameter Estimation of Ordinary Differential Equation Models

Ordinary differential equation (ODE) models are widely used to describe biochemical processes, since they effectively represent mass action kinetics. Optimization-based calibration of ODE models on experimental data can be challenging, even for low-dimensional problems. However, reliable model calibration is a prerequisite for uncertainty analysis, model comparison, and biological interpretation. Multiple hypotheses have been advanced to explain why optimization based calibration of biochemical models is challenging, but there are few comprehensive studies that test these hypotheses, likely because tools for performing such studies are also lacking. We implemented an established trust-region method as a modular Python framework (fides) to enable systematic comparison of different approaches to ODE model calibration involving various Hessian approximation schemes. We evaluated fides on a set of benchmark problems for which real experimental data are available. Unexpectedly, we observed high variability in optimizer performance among different implementations of the same algorithm. Overall, fides performed most reliably and efficiently. Our investigation of possible sources of poor optimizer performance identified drawbacks in the widely used Gauss-Newton, BFGS and SR1 Hessian approximations. We address these drawbacks by proposing a novel hybrid Hessian approximation scheme that enhances optimizer performance and outperforms existing hybrid approaches. We expect fides to be broadly useful for ODE constrained optimization problems and to enable future methods development. Availabilityfides is published under the permissive BSD-3-Clause license with source code publicly available at https://github.com/fides-dev/fides. Citeable releases are archived on Zenodo. Code to reproduce results presented in this manuscript is available at https://github.com/fides-dev/fides-benchmark.

systems biology↗

bioRxiv2021

Cancer patient survival can be accurately parameterized, revealing time-dependent therapeutic effects and doubling the precision of small trials

Individual participant data (IPD) from completed oncology clinical trials are a valuable but rarely available source of information. A lack of minable survival distributions has made it difficult to identify factors determining the success and failure of clinical trials and improve trial design. We imputed survival IPD from [~]500 arms of phase III oncology trials (representing [~]220,000 events) and found that they are well fit by a two-parameter Weibull distribution. This makes it possible to use parametric statistics to substantially increase trial precision with small patient cohorts typical of phase I or II trials. For example, a 50-person trial parameterized using Weibull distributions is as precise as a 90-person trial evaluated using traditional statistics. Mining IPD also showed that frequent violations of the proportional hazards assumption, particularly in trials of immune checkpoint inhibitors (ICIs), arise from time-dependent therapeutic effects and hazard ratios. Thus, the duration of ICI trials has an underappreciated impact on the likelihood of their success.

cancer biology↗

bioRxiv2021

Temporal and spatial topography of cell proliferation in cancer

Proliferation is a fundamental trait of cancer cells but is poorly characterized in tumors by classical histologic methods. We use multiplexed tissue imaging to quantify the abundance of multiple cell cycle regulating proteins at single-cell level and develop robust multivariate proliferation metrics. Across cancers, the proliferative architecture is organized at two distinct spatial scales: large domains, and local niches enriched for specific immune lineages. A subset of tumor cells express cell cycle regulators in canonical patterns consistent with unrestrained proliferation, a phenomenon we refer to as "cell cycle coherence". By contrast, the cell cycles of other tumor cell populations are skewed toward a specific phase or characterized by non-canonical (incoherent) marker combinations. Coherence varies across space, with changes in oncogene activity, and with therapeutic intervention, and is associated with aggressive behavior. Multivariate measures capture clinically significant features of cancer proliferation, a fundamental step in enabling more precise use of anti-cancer therapies.

systems biology↗

bioRxiv2021

Stitching and registering highly multiplexed whole slide images of tissues and tumors using ASHLAR software

MotivationStitching microscope images into a mosaic is an essential step in the analysis and visualization of large biological specimens, particularly human and animal tissues. Recent approaches to highly-multiplexed imaging generate high-plex data from sequential rounds of lower-plex imaging. These multiplexed imaging methods promise to yield precise molecular single-cell data and information on cellular neighborhoods and tissue architecture. However, attaining mosaic images with single-cell accuracy requires robust image stitching and image registration capabilities that are not met by existing methods. ResultsWe describe the development and testing of ASHLAR, a Python tool for coordinated stitching and registration of 103 or more individual multiplexed images to generate accurate whole-slide mosaics. ASHLAR reads image formats from most commercial microscopes and slide scanners, and we show that it performs better than existing open source and commercial software. ASHLAR outputs standard OME-TIFF images that are ready for analysis by other open-source tools and recently developed image analysis pipelines. Availability and implementationASHLAR is written in Python and available under an MIT license at https://github.com/labsyspharm/ashlar. An informational website with user guides and test data is available at https://labsyspharm.github.io/ashlar/.

bioinformatics↗

bioRxiv2021

UnMICST: Deep learning with real augmentation for robust segmentation of highly multiplexed images of human tissues

Newly developed technologies have made it feasible to routinely collect highly multiplexed (20-60 channel) images at subcellular resolution from human tissues for research and diagnostic purposes. Extracting single cell data from such images requires efficient and accurate image segmentation, a challenging problem that has recently benefited from the use of deep learning. In this paper, we demonstrate two approaches to improving tissue segmentation that are applicable to multiple deep learning frameworks. The first uses "real augmentations" that comprise defocused and saturated image data collected on the same instruments as the actual data; using real augmentation improves model accuracy to a significantly greater degree than computational augmentation (Gaussian blurring). The second involves imaging the nuclear envelope to better identify nuclear outlines. The two approaches cumulatively and substantially improve segmentation on a wide range of tissue types and provide a set of improved segmentation models. We speculate that the use of real augmentations may have applications in image processing outside of microscopy.

bioinformatics↗

bioRxiv2021

Multiplexed 3D atlas of state transitions and immune interactions in colorectal cancer

Advanced solid cancers are complex assemblies of tumor, immune, and stromal cells characterized by high intratumoral variation. We use highly multiplexed tissue imaging, 3D reconstruction, spatial statistics, and machine learning to identify cell types and states underlying morphological features of known diagnostic and prognostic significance in colorectal cancer. Quantitation of these features in high-plex marker space reveals recurrent transitions from one tumor morphology to the next, some of which are coincident with long-range gradients in the expression of oncogenes and epigenetic regulators. At the tumor invasive margin, where tumor, normal, and immune cells compete, T-cell suppression involves multiple cell types and 3D imaging shows that seemingly localized 2D features such as tertiary lymphoid structures are commonly interconnected and have graded molecular properties. Thus, while cancer genetics emphasizes the importance of discrete changes in tumor state, whole-specimen imaging reveals large-scale morphological and molecular gradients analogous to those in developing tissues.

cancer biology↗

bioRxiv2021

Micro-region transcriptomics of fixed human tissue using Pick-Seq

Spatial transcriptomics and multiplexed imaging are complementary methods for studying tissue biology and disease. Recently developed spatial transcriptomic methods use fresh-frozen specimens but most diagnostic specimens, clinical trials, and tissue archives rely on formaldehyde-fixed tissue. Here we describe the Pick-Seq method for deep spatial transcriptional profiling of fixed tissue. Pick-Seq is a form of micro-region sequencing in which small regions of tissue, containing 5-20 cells, are mechanically isolated on a microscope and then sequenced. We demonstrate the use of Pick-Seq with several different fixed and frozen human specimens. Application of Pick-Seq to a human melanoma with complex histology reveals significant differences in transcriptional programs associated with tumor invasion, proliferation, and immuno-editing. Parallel imaging confirms changes in immuno-phenotypes and cancer cell states. This work demonstrates the ability of Pick-Seq to generate deep spatial transcriptomic data from fixed and archival tissue with multiplexed imaging in parallel.

cancer biology↗

bioRxiv2021

MCMICRO: A scalable, modular image-processing pipeline for multiplexed tissue imaging

Highly multiplexed tissue imaging makes molecular analysis of single cells possible in a preserved spatial context. However, reproducible analysis of the underlying data poses a substantial computational challenge. Here we describe a modular and open-source computational pipeline (MCMICRO) for performing the sequential steps needed to transform large, multi-channel whole slide images into single-cell data. We demonstrate use of MCMICRO on images of different tissues and tumors acquired using multiple imaging platforms, thereby providing a solid foundation for the continued development of tissue imaging software.

bioinformatics↗

bioRxiv2020

Proteomic profiling of breast cancer cell lines and models

We performed quantitative proteomics on 61 human-derived breast cancer cell lines to a depth of ~13,000 proteins. The resulting high-throughput datasets were assessed for quality and reproducibility. We used the datasets to identify and characterize the subtypes of breast cancer and showed that they conform to known transcriptional subtypes, revealing that molecular subtypes are preserved even in under-sampled protein feature sets. All datasets are freely available as public resources on the LINCS portal. We anticipate that these datasets, either in isolation or in combination with complimentary measurements such as genomics, transcriptomics and phosphoproteomics, can be mined for the purpose of predicting drug response, informing cell line specific context in models of signalling pathways, and identifying markers of sensitivity or resistance to therapeutics.

cancer biology↗

bioRxiv2020

Three-dimensional spatial transcriptomics uncovers cell type dynamics in the rheumatoid arthritis synovium

The inflamed rheumatic joint is a highly heterogeneous and complex tissue with dynamic recruitment and expansion of multiple cell types that interact in multifaceted ways within a localized area. Rheumatoid arthritis synovium has primarily been studied either by immunostaining or by molecular profiling after tissue homogenization. Here, we use Spatial Transcriptomics to study local cellular interactions at the site of chronic synovial inflammation. We report comprehensive spatial RNA-seq data coupled to quantitative and cell type-specific chemokine-driven dynamics at and around organized structures of infiltrating leukocyte cells in the synovium.

genomics↗

bioRxiv2020

Channel Embedding for Informative Protein Identification from Highly Multiplexed Images

Interest is growing rapidly in using deep learning to classify biomedical images, and interpreting these deep-learned models is necessary for life-critical decisions and scientific discovery. Effective interpretation techniques accelerate biomarker discovery and provide new insights into the etiology, diagnosis, and treatment of disease. Most interpretation techniques aim to discover spatially-salient regions within images, but few techniques consider imagery with multiple channels of information. For instance, highly multiplexed tumor and tissue images have 30-100 channels and require interpretation methods that work across many channels to provide deep molecular insights. We propose a novel channel embedding method that extracts features from each channel. We then use these features to train a classifier for prediction. Using this channel embedding, we apply an interpretation method to rank the most discriminative channels. To validate our approach, we conduct an ablation study on a synthetic dataset. Moreover, we demonstrate that our method aligns with biological findings on highly multiplexed images of breast cancer cells while outperforming baseline pipelines.

pathology↗

bioRxiv2020

Biomarker-guided treatment strategies for ovarian cancer identified from a heterogeneous panel of patient-derived tumor xenografts

Advanced ovarian cancers are a leading cause of cancer-related death in women. Such cancers are currently treated with surgery and chemotherapy which is often temporarily successful but exhibits a high rate of relapse after which treatment options are few. Here we assess the responses of a panel of patient-derived ovarian cancer xenografts (PDXs) to 19 mono and combination therapies, including small molecules and antibody-drug conjugates. The PDX panel aimed to mimic the heterogeneity of disease observed in patients, and exhibited a distribution of responsiveness to standard of care chemotherapy similar to human clinical data. Three monotherapies and one drug combination were found to be active in different subsets of PDXs. By analyzing gene expression data we identified gene expression biomarkers predictive of responsiveness to each of three novel targeted therapy regimens. While no single treatment had as high a response rate as chemotherapy, nearly 90% of PDXs were eligible for and responded to at least one biomarker-guided treatment, including tumors resistant to standard chemotherapy. Biomarker frequency was similar in human patients, suggesting the possibility of a new therapeutic approach to ovarian cancer and demonstrating the potential power of PDX-based trials in broadening the reach of precision cancer medicine.

cancer biology↗

bioRxiv2019

Assembling a corpus of phosphoproteomic annotations using ProtMapper to normalize site information from databases and text mining

Protein phosphorylation regulates numerous cellular processes and is highly studied in biology.However, the analysis of phosphoproteomic datasets remains challenging due to limited information on upstream regulators of phosphosites, which is fragmented across multiple curated databases and unstructured literature. When aggregating information on phosphosites from six databases and three text mining systems, we found that a substantial proportion of phosphosites were mentioned at residue positions not matching the reference sequence. These errors were often attributable to the use of residue numbers from non-canonical protein isoforms, mouse or rat proteins, or post-translationally processed proteins. Non-canonical site numbering is also prevalent in mass spectrometry datasets from large-scale efforts such as the Clinical Proteomic Tumor Analysis Consortium (CPTAC). To address these issues, we developed ProtMapper, an open-source Python tool that automatically normalizes site positions to human protein reference sequences. We used ProtMapper coupled with the INDRA knowledge assembly system to create a corpus of 37,028 regulatory annotations for 16,332 sites - to our knowledge, the most comprehensive corpus of literature-derived information about phosphosite regulation currently available. This work highlights how automated phosphosite normalization coupled to text mining and knowledge assembly allows researchers to leverage phosphosite information that exists within the scientific literature.

bioinformatics↗

bioRxiv2019

Sporadic ERK pulses drive non-genetic resistance in drug-adapted BRAFV600E melanoma cells

Anti-cancer drugs commonly target signal transduction proteins activated by mutation. In patients with BRAFV600E melanoma, small molecule RAF and MEK kinase inhibitors cause dramatic but often transient tumor regression. Emerging evidence suggests that cancer cells adapting by non-genetic mechanisms constitute a reservoir for the development of drug-resistant tumors. Here, we show that few hours after exposure to RAF/MEK inhibitors, BRAFV600E melanomas undergo adaptive changes involving disruption of negative feedback and sporadic pulsatile reactivation of the MAPK pathway, so that MAPK activity is transiently high enough in some cells to drive proliferation. Quantitative proteomics and computational modeling show that pulsatile MAPK reactivation is possible due to the co-existence in cells of two MAPK cascades: one driven by BRAFV600E that is drug-sensitive and a second driven by receptors that is drug-resistant. Paradoxically, this may account both for the frequent emergence of drug resistance and for the tolerability of RAF/MEK therapy in patients.

cancer biology↗

bioRxiv2019

GeneWalk identifies relevant gene functions for a biological context using network representation learning

The primary bottleneck in high-throughput genomics experiments is identifying the most important genes and their relevant functions from a list of gene hits. Existing methods such as Gene Ontology (GO) enrichment analysis provide insight at the gene set level. For individual genes, GO annotations are static and biological context can only be added by manual literature searches. Here, we introduce GeneWalk (github.com/churchmanlab/genewalk), a method that identifies individual genes and their relevant functions under a particular experimental condition. After automatic assembly of an experiment-specific gene regulatory network, GeneWalk quantifies the similarity between vector representations of each gene and its GO annotations through representation learning, yielding annotation significance scores that reflect their functional relevance for the experimental context. We demonstrate the use of GeneWalk analysis of RNA-seq and nascent transcriptome (NET-seq) data from human cells and mouse brains, validating the methodology. By performing gene- and condition-specific functional analysis that converts a list of genes into data-driven hypotheses, GeneWalk accelerates the interpretation of high-throughput genetics experiments.

bioinformatics↗

bioRxiv2019

Inferring reaction network structure from single-cell, multiplex data, using toric systems theory

The goal of many single-cell studies on eukaryotic cells is to gain insight into the biochemical reactions that control cell fate and state. In this paper we introduce the concept of effective stoichiometric space (ESS) to guide the reconstruction of biochemical networks from multiplexed, fixed time-point, single-cell data. In contrast to methods based solely on statistical models of data, the ESS method leverages the power of the geometric theory of toric varieties to begin unraveling the structure of chemical reaction networks (CRN). This application of toric theory enables a data-driven mapping of covariance relationships in single cell measurements into stoichiometric information, one in which each cell subpopulation has its associated ESS interpreted in terms of CRN theory. In the development of ESS we reframe certain aspects of the theory of CRN to better match data analysis. As an application of our approach we process cytomery- and image-based single-cell datasets and identify differences in cells treated with kinase inhibitors. Our approach is directly applicable to data acquired using readily accessible experimental methods such as Fluorescence Activated Cell Sorting (FACS) and multiplex immunofluorescence.\n\nAuthor summaryWe introduce a new notion, which we call the effective stoichiometric space (ESS), that elucidates network structure from the covariances of single-cell multiplexed data. The ESS approach differs from methods that are based on purely statistical models of data: it allows a completely new and data-driven translation of the theory of toric varieties in geometry and specifically their role in chemical reaction networks (CRN). In the process, we reframe certain aspects of the theory of CRN. As illustrations of our approach, we find stoichiometry in different single-cell datasets, and pinpoint dose-dependence of network perturbations in drug-treated cells.

systems biology↗

bioRxiv2019

Opposing immune and genetic forces shape oncogenic programs in synovial sarcoma

Synovial sarcoma is an aggressive mesenchymal neoplasm, driven by the SS18-SSX fusion, and characterized by immunogenic antigens expression and exceptionally low T cell infiltration levels. To study the cancer-immune interplay in this disease, we profiled 16,872 cells from 12 human synovial sarcoma tumors using single-cell RNA-sequencing (scRNA-Seq). Synovial sarcoma manifests antitumor immunity, high cellular plasticity and a core oncogenic program, which is predictive of low immune levels and poor clinical outcomes. Using genetic and pharmacological perturbations, we demonstrate that the program is controlled by the SS18-SSX driver and repressed by cytokines secreted by macrophages and T cells in the tumor microenvironment. Network modeling predicted that SS18-SSX promotes the program through HDAC1 and CDK6. Indeed, the combination of HDAC and CDK4/6 inhibitors represses the program, induces immunogenic cell states, and selectively targets synovial sarcoma cells. Our study demonstrates that immune evasion, cellular plasticity, and cell cycle are co-regulated and can be co-targeted in synovial sarcoma and potentially in other malignancies.

cancer biology↗