Imputing Single-Cell Protein Abundance in Multiplex Tissue Imaging.

Multiplex tissue imaging are a collection of increasingly popular single-cell spatial proteomics and transcriptomics assays for characterizing biological tissues both compositionally and spatially. However, several technical issues limit the utility of multiplex tissue imaging, including the limited number of RNAs and proteins that can be assayed, tissue loss, and protein probe failure. In this work, we demonstrate how machine learning methods can address these limitations by imputing protein abundance at the single-cell level using multiplex tissue imaging datasets from a breast cancer cohort. We first compared machine learning methods' strengths and weaknesses for imputing single-cell protein abundance. Machine learning methods used in this work include regularized linear regression, gradient-boosted regression trees, and deep learning autoencoders. We also incorporated cellular spatial information to improve imputation performance. Using machine learning, single-cell protein expression can be imputed with mean absolute error ranging between 0.05-0.3 on a [0,1] scale. Our results demonstrate (1) the feasibility of imputing single-cell abundance levels for many proteins using machine learning to overcome the technical constraints of multiplex tissue imaging and (2) how including cellular spatial information can substantially enhance imputation results.

Addressing persistent challenges in digital image analysis of cancerous tissues.

The National Cancer Institute (NCI) supports many research programs and consortia, many of which use imaging as a major modality for characterizing cancerous tissue. A trans-consortia Image Analysis Working Group (IAWG) was established in 2019 with a mission to disseminate imaging-related work and foster collaborations. In 2022, the IAWG held a virtual hackathon focused on addressing challenges of analyzing high dimensional datasets from fixed cancerous tissues. Standard image processing techniques have automated feature extraction, but the next generation of imaging data requires more advanced methods to fully utilize the available information. In this perspective, we discuss current limitations of the automated analysis of multiplexed tissue images, the first steps toward deeper understanding of these limitations, what possible solutions have been developed, any new or refined approaches that were developed during the Image Analysis Hackathon 2022, and where further effort is required. The outstanding problems addressed in the hackathon fell into three main themes: 1) challenges to cell type classification and assessment, 2) translation and visual representation of spatial aspects of high dimensional data, and 3) scaling digital image analyses to large (multi-TB) datasets. We describe the rationale for each specific challenge and the progress made toward addressing it during the hackathon. We also suggest areas that would benefit from more focus and offer insight into broader challenges that the community will need to address as new technologies are developed and integrated into the broad range of image-based modalities and analytical resources already in use within the cancer research community.

An omic and multidimensional spatial atlas from serial biopsies of an evolving metastatic breast cancer.

Mechanisms of therapeutic resistance and vulnerability evolve in metastatic cancers as tumor cells and extrinsic microenvironmental influences change during treatment. To support the development of methods for identifying these mechanisms in individual people, here we present an omic and multidimensional spatial (OMS) atlas generated from four serial biopsies of an individual with metastatic breast cancer during 3.5 years of therapy. This resource links detailed, longitudinal clinical metadata that includes treatment times and doses, anatomic imaging, and blood-based response measurements to clinical and exploratory analyses, which includes comprehensive DNA, RNA, and protein profiles; images of multiplexed immunostaining; and 2- and 3-dimensional scanning electron micrographs. These data report aspects of heterogeneity and evolution of the cancer genome, signaling pathways, immune microenvironment, cellular composition and organization, and ultrastructure. We present illustrative examples of how integrative analyses of these data reveal potential mechanisms of response and resistance and suggest novel therapeutic vulnerabilities.

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

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

Multiomics analysis of serial PARP inhibitor treated metastatic TNBC inform on rational combination therapies.

In a pilot study, we evaluated the feasibility of real-time deep analysis of serial tumor samples from triple negative breast cancer patients to identify mechanisms of resistance and treatment opportunities as they emerge under therapeutic stress engendered by poly-ADP-ribose polymerase (PARP) inhibitors (PARPi). In a BRCA-mutant basal breast cancer exceptional long-term survivor, a striking tumor destruction was accompanied by a marked infiltration of immune cells containing CD8 effector cells, consistent with pre-clinical evidence for association between STING mediated immune activation and benefit from PARPi and immunotherapy. Tumor cells in the exceptional responder underwent extensive protein network rewiring in response to PARP inhibition. In contrast, there were minimal changes in the ecosystem of a luminal androgen receptor rapid progressor, likely due to indifference to the effects of PARP inhibition. Together, identification of PARPi-induced emergent changes could be used to select patient specific combination therapies, based on tumor and immune state changes.

Phase Ib Dose Expansion and Translational Analyses of Olaparib in Combination with Capivasertib in Recurrent Endometrial, Triple-Negative Breast, and Ovarian Cancer.

PURPOSE: On the basis of strong preclinical rationale, we sought to confirm recommended phase II dose (RP2D) for olaparib, a PARP inhibitor, combined with the AKT inhibitor capivasertib and assess molecular markers of response and resistance. PATIENTS AND METHODS: We performed a safety lead-in followed by expansion in endometrial, triple-negative breast, ovarian, fallopian tube, or peritoneal cancer. Olaparib 300 mg orally twice daily and capivasertib orally twice daily on a 4-day on 3-day off schedule was evaluated. Two dose levels (DL) of capivasertib were planned: 400 mg (DL1) and 320 mg (DL-1). Patients underwent biopsies at baseline and 28 days. RESULTS: A total of 38 patients were enrolled. Seven (18%) had germline BRCA1/2 mutations. The first 2 patients on DL1 experienced dose-limiting toxicities (DLT) of diarrhea and vomiting. No DLTs were observed on DL-1 (n = 6); therefore, DL1 was reexplored (n = 6) with no DLTs, confirming DL1 as RP2D. Most common treatment-related grade 3/4 adverse events were anemia (23.7%) and leukopenia (10.5%). Of 32 evaluable subjects, 6 (19%) had partial response (PR); PR rate was 44.4% in endometrial cancer. Seven (22%) additional patients had stable disease greater than 4 months. Tumor analysis demonstrated strong correlations between response and immune activity, cell-cycle alterations, and DNA damage response. Therapy resistance was associated with receptor tyrosine kinase and RAS-MAPK pathway activity, metabolism, and epigenetics. CONCLUSIONS: The combination of olaparib and capivasertib is associated to no serious adverse events and demonstrates durable activity in ovarian, endometrial, and breast cancers, with promising responses in endometrial cancer. Importantly, tumor samples acquired pre- and on-therapy can help predict patient benefit.

Galaxy-ML: An accessible, reproducible, and scalable machine learning toolkit for biomedicine.

Supervised machine learning is an essential but difficult to use approach in biomedical data analysis. The Galaxy-ML toolkit (https://galaxyproject.org/community/machine-learning/) makes supervised machine learning more accessible to biomedical scientists by enabling them to perform end-to-end reproducible machine learning analyses at large scale using only a web browser. Galaxy-ML extends Galaxy (https://galaxyproject.org), a biomedical computational workbench used by tens of thousands of scientists across the world, with a suite of tools for all aspects of supervised machine learning.

A community challenge to evaluate RNA-seq, fusion detection, and isoform quantification methods for cancer discovery.

The accurate identification and quantitation of RNA isoforms present in the cancer transcriptome is key for analyses ranging from the inference of the impacts of somatic variants to pathway analysis to biomarker development and subtype discovery. The ICGC-TCGA DREAM Somatic Mutation Calling in RNA (SMC-RNA) challenge was a crowd-sourced effort to benchmark methods for RNA isoform quantification and fusion detection from bulk cancer RNA sequencing (RNA-seq) data. It concluded in 2018 with a comparison of 77 fusion detection entries and 65 isoform quantification entries on 51 synthetic tumors and 32 cell lines with spiked-in fusion constructs. We report the entries used to build this benchmark, the leaderboard results, and the experimental features associated with the accurate prediction of RNA species. This challenge required submissions to be in the form of containerized workflows, meaning each of the entries described is easily reusable through CWL and Docker containers at https://github.com/SMC-RNA-challenge. A record of this paper's transparent peer review process is included in the supplemental information.

Characterizing advanced breast cancer heterogeneity and treatment resistance through serial biopsies and comprehensive analytics.

Molecular heterogeneity in metastatic breast cancer presents multiple clinical challenges in accurately characterizing and treating the disease. Current diagnostic approaches offer limited ability to assess heterogeneity that exists among multiple metastatic lesions throughout the treatment course. We developed a precision oncology platform that combines serial biopsies, multi-omic analysis, longitudinal patient monitoring, and molecular tumor boards, with the goal of improving cancer management through enhanced understanding of the entire cancer ecosystem within each patient. We describe this integrative approach using comprehensive analytics generated from serial-biopsied lesions in a metastatic breast cancer patient. The serial biopsies identified remarkable heterogeneity among metastatic lesions that presented clinically as discordance in receptor status and genomic alterations with mixed treatment response. Based on our study, we highlight clinical scenarios, such as rapid progression or mixed response, that indicate consideration for repeat biopsies to evaluate intermetastatic heterogeneity (IMH), with the objective of refining targeted therapy. We present a framework for understanding the clinical significance of heterogeneity in breast cancer between metastatic lesions utilizing multi-omic analyses of serial biopsies and its implication for effective personalized treatment.

Phytopathogenic Rhodococcus Have Diverse Plasmids With Few Conserved Virulence Functions.

Rhodococcus is a genus of Gram-positive bacteria with species that can cause growth deformations to a large number of plant species. This ability to cause disease is hypothesized to be dependent on a cluster of three gene loci on an almost 200 kb-sized linear plasmid. To reevaluate the roles of some of the genes in pathogenicity, we constructed and characterized deletion mutants of fasR and four fas genes. Findings confirmed that fasR, which encodes a putative transcriptional regulator, is necessary for pathogenesis. However, three of the fas genes, implicated in the metabolism of plant growth promoting cytokinins, are dispensable for the ability of the pathogen to cause disease. We also used long-read sequencing technology to generate high quality genome sequences for two phytopathogenic strains in which virulence genes are diverged in sequence and/or hypothesized to have recombined into the chromosome. Surprisingly, findings showed that the two strains carry extremely diverse virulence plasmids. Ortholog clustering identified only 12 genes present on all three virulence plasmids. Rhodococcus requires a small number of horizontally acquired traits to be pathogenic and the transmission of the corresponding genes, via recombination and conjugation, has the potential to rapidly diversify plasmids and bacterial populations.

Reproducible biomedical benchmarking in the cloud: lessons from crowd-sourced data challenges.

Challenges are achieving broad acceptance for addressing many biomedical questions and enabling tool assessment. But ensuring that the methods evaluated are reproducible and reusable is complicated by the diversity of software architectures, input and output file formats, and computing environments. To mitigate these problems, some challenges have leveraged new virtualization and compute methods, requiring participants to submit cloud-ready software packages. We review recent data challenges with innovative approaches to model reproducibility and data sharing, and outline key lessons for improving quantitative biomedical data analysis through crowd-sourced benchmarking challenges.

Preconception Carrier Screening by Genome Sequencing: Results from the Clinical Laboratory.

Advances in sequencing technologies permit the analysis of a larger selection of genes for preconception carrier screening. The study was designed as a sequential carrier screen using genome sequencing to analyze 728 gene-disorder pairs for carrier and medically actionable conditions in 131 women and their partners (n = 71) who were planning a pregnancy. We report here on the clinical laboratory results from this expanded carrier screening program. Variants were filtered and classified using the latest American College of Medical Genetics and Genomics (ACMG) guideline; only pathogenic and likely pathogenic variants were confirmed by orthologous methods before being reported. Novel missense variants were classified as variants of uncertain significance. We reported 304 variants in 202 participants. Twelve carrier couples (12/71 couples tested) were identified for common conditions; eight were carriers for hereditary hemochromatosis. Although both known and novel variants were reported, 48% of all reported variants were missense. For novel splice-site variants, RNA-splicing assays were performed to aid in classification. We reported ten copy-number variants and five variants in non-coding regions. One novel variant was reported in F8, associated with hemophilia A; prenatal testing showed that the male fetus harbored this variant and the neonate suffered a life-threatening hemorrhage which was anticipated and appropriately managed. Moreover, 3% of participants had variants that were medically actionable. Compared with targeted mutation screening, genome sequencing improves the sensitivity of detecting clinically significant variants. While certain novel variant interpretation remains challenging, the ACMG guidelines are useful to classify variants in a healthy population.

Evolutionary transitions between beneficial and phytopathogenic Rhodococcus challenge disease management.

Understanding how bacteria affect plant health is crucial for developing sustainable crop production systems. We coupled ecological sampling and genome sequencing to characterize the population genetic history of Rhodococcus and the distribution patterns of virulence plasmids in isolates from nurseries. Analysis of chromosome sequences shows that plants host multiple lineages of Rhodococcus, and suggested that these bacteria are transmitted due to independent introductions, reservoir populations, and point source outbreaks. We demonstrate that isolates lacking virulence genes promote beneficial plant growth, and that the acquisition of a virulence plasmid is sufficient to transition beneficial symbionts to phytopathogens. This evolutionary transition, along with the distribution patterns of plasmids, reveals the impact of horizontal gene transfer in rapidly generating new pathogenic lineages and provides an alternative explanation for pathogen transmission patterns. Results also uncovered a misdiagnosed epidemic that implicated beneficial Rhodococcus bacteria as pathogens of pistachio. The misdiagnosis perpetuated the unnecessary removal of trees and exacerbated economic losses.

The Pseudomonas syringae Type III Effector HopG1 Induces Actin Remodeling to Promote Symptom Development and Susceptibility during Infection.

The plant cytoskeleton underpins the function of a multitude of cellular mechanisms, including those associated with developmental- and stress-associated signaling processes. In recent years, the actin cytoskeleton has been demonstrated to play a key role in plant immune signaling, including a recent demonstration that pathogens target actin filaments to block plant defense and immunity. Herein, we quantified spatial changes in host actin filament organization after infection with Pseudomonas syringae pv. tomato DC3000 (Pst DC3000), demonstrating that the type-III effector HopG1 is required for pathogen-induced changes to actin filament architecture and host disease symptom development during infection. Using a suite of pathogen effector deletion constructs, coupled with high-resolution microscopy, we found that deletion of hopG1 from Pst DC3000 resulted in a reduction in actin bundling and a concomitant increase in the density of filament arrays in Arabidopsis, both of which correlate with host disease symptom development. As a mechanism underpinning this activity, we further show that the HopG1 effector interacts with an Arabidopsis mitochondrial-localized kinesin motor protein. Kinesin mutant plants show reduced disease symptoms after pathogen infection, which can be complemented by actin-modifying agents. In total, our results support a model in which HopG1 induces changes in the organization of the actin cytoskeleton as part of its virulence function in promoting disease symptom development.

Genome Sequencing and Transposon Mutagenesis of Burkholderia seminalis TC3.4.2R3 Identify Genes Contributing to Suppression of Orchid Necrosis Caused by B. gladioli.

From a screen of 36 plant-associated strains of Burkholderia spp., we identified 24 strains that suppressed leaf and pseudobulb necrosis of orchid caused by B. gladioli. To gain insights into the mechanisms of disease suppression, we generated a draft genome sequence from one suppressive strain, TC3.4.2R3. The genome is an estimated 7.67 megabases in size, with three replicons, two chromosomes, and the plasmid pC3. Using a combination of multilocus sequence analysis and phylogenomics, we identified TC3.4.2R3 as B. seminalis, a species within the Burkholderia cepacia complex that includes opportunistic human pathogens and environmental strains. We generated and screened a library of 3,840 transposon mutants of strain TC3.4.2R3 on orchid leaves to identify genes contributing to plant disease suppression. Twelve mutants deficient in suppression of leaf necrosis were selected and the transposon insertions were mapped to eight loci. One gene is in a wcb cluster that is related to synthesis of extracellular polysaccharide, a key determinant in bacterial-host interactions in other systems, and the other seven are highly conserved among Burkholderia spp. The fundamental information developed in this study will serve as a resource for future research aiming to identify mechanisms contributing to biological control.

Use of whole genome sequences to develop a molecular phylogenetic framework for Rhodococcus fascians and the Rhodococcus genus.

The accurate diagnosis of diseases caused by pathogenic bacteria requires a stable species classification. Rhodococcus fascians is the only documented member of its ill-defined genus that is capable of causing disease on a wide range of agriculturally important plants. Comparisons of genome sequences generated from isolates of Rhodococcus associated with diseased plants revealed a level of genetic diversity consistent with them representing multiple species. To test this, we generated a tree based on more than 1700 homologous sequences from plant-associated isolates of Rhodococcus, and obtained support from additional approaches that measure and cluster based on genome similarities. Results were consistent in supporting the definition of new Rhodococcus species within clades containing phytopathogenic members. We also used the genome sequences, along with other rhodococcal genome sequences to construct a molecular phylogenetic tree as a framework for resolving the Rhodococcus genus. Results indicated that Rhodococcus has the potential for having 20 species and also confirmed a need to revisit the taxonomic groupings within Rhodococcus.

Analysis of genome sequences from plant pathogenic Rhodococcus reveals genetic novelties in virulence loci.

Members of Gram-positive Actinobacteria cause economically important diseases to plants. Within the Rhodococcus genus, some members can cause growth deformities and persist as pathogens on a wide range of host plants. The current model predicts that phytopathogenic isolates require a cluster of three loci present on a linear plasmid, with the fas operon central to virulence. The Fas proteins synthesize, modify, and activate a mixture of growth regulating cytokinins, which cause a hormonal imbalance in plants, resulting in abnormal growth. We sequenced and compared the genomes of 20 isolates of Rhodococcus to gain insights into the mechanisms and evolution of virulence in these bacteria. Horizontal gene transfer was identified as critical but limited in the scale of virulence evolution, as few loci are conserved and exclusive to phytopathogenic isolates. Although the fas operon is present in most phytopathogenic isolates, it is absent from phytopathogenic isolate A21d2. Instead, this isolate has a horizontally acquired gene chimera that encodes a novel fusion protein with isopentyltransferase and phosphoribohydrolase domains, predicted to be capable of catalyzing and activating cytokinins, respectively. Cytokinin profiling of the archetypal D188 isolate revealed only one activate cytokinin type that was specifically synthesized in a fas-dependent manner. These results suggest that only the isopentenyladenine cytokinin type is synthesized and necessary for Rhodococcus phytopathogenicity, which is not consistent with the extant model stating that a mixture of cytokinins is necessary for Rhodococcus to cause leafy gall symptoms. In all, data indicate that only four horizontally acquired functions are sufficient to confer the trait of phytopathogenicity to members of the genetically diverse clade of Rhodococcus.

The ABCs and 123s of bacterial secretion systems in plant pathogenesis.

Bacteria have many export and secretion systems that translocate cargo into and across biological membranes. Seven secretion systems contribute to pathogenicity by translocating proteinaceous cargos that can be released into the extracellular milieu or directly into recipient cells. In this review, we describe these secretion systems and how their complexities and functions reflect differences in the destinations, states, functions, and sizes of the translocated cargos as well as the architecture of the bacterial cell envelope. We examine the secretion systems from the perspective of pathogenic bacteria that proliferate within plant tissues and highlight examples of translocated proteins that contribute to the infection and disease of plant hosts.

Mutualistic co-evolution of type III effector genes in Sinorhizobium fredii and Bradyrhizobium japonicum.

Two diametric paradigms have been proposed to model the molecular co-evolution of microbial mutualists and their eukaryotic hosts. In one, mutualist and host exhibit an antagonistic arms race and each partner evolves rapidly to maximize their own fitness from the interaction at potential expense of the other. In the opposing model, conflicts between mutualist and host are largely resolved and the interaction is characterized by evolutionary stasis. We tested these opposing frameworks in two lineages of mutualistic rhizobia, Sinorhizobium fredii and Bradyrhizobium japonicum. To examine genes demonstrably important for host-interactions we coupled the mining of genome sequences to a comprehensive functional screen for type III effector genes, which are necessary for many Gram-negative pathogens to infect their hosts. We demonstrate that the rhizobial type III effector genes exhibit a surprisingly high degree of conservation in content and sequence that is in contrast to those of a well characterized plant pathogenic species. This type III effector gene conservation is particularly striking in the context of the relatively high genome-wide diversity of rhizobia. The evolution of rhizobial type III effectors is inconsistent with the molecular arms race paradigm. Instead, our results reveal that these loci are relatively static in rhizobial lineages and suggest that fitness conflicts between rhizobia mutualists and their host plants have been largely resolved.