A Cancer Biologist's Primer on Machine Learning Applications in High-Dimensional Cytometry

del Hierro A, H-Vázquez J, Cuesta-Sancho S, Bernardo D. State-of-the-art cytometry in the search of novel biomarkers in digestive cancers

Despite that colorectal and liver cancer are among the most prevalent tumours in the world, the identification of non-invasive biomarkers to aid on their diagnose and subsequent prognosis is a current unmet need that would diminish both their incidence and mortality rates. In this context, conventional flow cytometry has been widely used in the screening of biomarkers with clinical utility in other malignant processes like leukaemia or lymphoma. Therefore, in this review, we will focus on how advanced cytometry panels covering over 40 parameters can be applied on the study of the immune system from patients with colorectal and hepatocellular carcinoma and how that can be used on the search of novel biomarkers to aid or diagnose, prognosis, and even predict clinical response to different treatments. In addition, these multiparametric and unbiased approaches can also provide novel insights into the specific immunopathogenic mechanisms governing these malignant diseases, hence potentially unravelling novel targets to perform immunotherapy or identify novel mechanisms, rendering the development of novel treatments. As a consequence, computational cytometry approaches are an emerging methodology for the early detection and predicting therapies for gastrointestinal cancers.

A comparison of RNA-Seq data preprocessing pipelines for transcriptomic predictions across independent studies

BACKGROUND

RNA sequencing combined with machine learning techniques has provided a modern approach to the molecular classification of cancer. Class predictors, reflecting the disease class, can be constructed for known tissue types using the gene expression measurements extracted from cancer patients. One challenge of current cancer predictors is that they often have suboptimal performance estimates when integrating molecular datasets generated from different labs. Often, the quality of the data is variable, procured differently, and contains unwanted noise hampering the ability of a predictive model to extract useful information. Data preprocessing methods can be applied in attempts to reduce these systematic variations and harmonize the datasets before they are used to build a machine learning model for resolving tissue of origins.

RESULTS

We aimed to investigate the impact of data preprocessing steps-focusing on normalization, batch effect correction, and data scaling-through trial and comparison. Our goal was to improve the cross-study predictions of tissue of origin for common cancers on large-scale RNA-Seq datasets derived from thousands of patients and over a dozen tumor types. The results showed that the choice of data preprocessing operations affected the performance of the associated classifier models constructed for tissue of origin predictions in cancer.

CONCLUSION

By using TCGA as a training set and applying data preprocessing methods, we demonstrated that batch effect correction improved performance measured by weighted F1-score in resolving tissue of origin against an independent GTEx test dataset. On the other hand, the use of data preprocessing operations worsened classification performance when the independent test dataset was aggregated from separate studies in ICGC and GEO. Therefore, based on our findings with these publicly available large-scale RNA-Seq datasets, the application of data preprocessing techniques to a machine learning pipeline is not always appropriate.

Flow Cytometry: The Next Revolution

Unmasking the subtleties of the immune system requires both a comprehensive knowledge base and the ability to interrogate that system with intimate sensitivity. That task, to a considerable extent, has been handled by an iterative expansion in flow cytometry methods, both in technological capability and also in accompanying advances in informatics. As the field of fluorescence-based cytomics matured, it reached a technological barrier at around 30 parameter analyses, which stalled the field until spectral flow cytometry created a fundamental transformation that will likely lead to the potential of 100 simultaneous parameter analyses within a few years. The simultaneous advance in informatics has now become a watershed moment for the field as it competes with mature systematic approaches such as genomics and proteomics, allowing cytomics to take a seat at the multi-omics table. In addition, recent technological advances try to combine the speed of flow systems with other detection methods, in addition to fluorescence alone, which will make flow-based instruments even more indispensable in any biological laboratory. This paper outlines current approaches in cell analysis and detection methods, discusses traditional and microfluidic sorting approaches as well as next-generation instruments, and provides an early look at future opportunities that are likely to arise.

Single-Cell Analysis and Classification according to Multiplexed Proteins via Microdroplet-Based Self-Driven Magnetic Surface-Enhanced Raman Spectroscopy Platforms Assisted with Machine Learning Algorithms

A microdroplet-based surface-enhanced Raman spectroscopy (microdroplet SERS) platform was constructed to envelop individual cells in microdroplets, followed by the SERS detection of their extracellular vesicle-proteins (EV-proteins) via the in-drop immunoassays by use of immunomagnetic beads (iMBs) and immuno-SERS tags (iSERS tags). A unique phenomenon is found that iMBs can start a spontaneous reorientation on the probed cell surface based on the electrostatic force-driven interfacial aggregation effect, which leads EV-proteins and iSERS tags to be gathered from a liquid phase to a cell membrane interface and significantly improves SERS sensitivity to the single-cell analysis level due to the formation of numbers of SERS hotspots. Three EV-proteins from two breast cancer cell lines were collected and further analyzed by machine learning algorithmic tools, which will be helpful for a deeper understanding of breast cancer subtypes from the view of EV-proteins.

tidytof: a user-friendly framework for scalable and reproducible high-dimensional cytometry data analysis

Summary

While many algorithms for analyzing high-dimensional cytometry data have now been developed, the software implementations of these algorithms remain highly customized-this means that exploring a dataset requires users to learn unique, often poorly interoperable package syntaxes for each step of data processing. To solve this problem, we developed {tidytof}, an open-source R package for analyzing high-dimensional cytometry data using the increasingly popular 'tidy data' interface.

Availability and implementation

{tidytof} is available at https://github.com/keyes-timothy/tidytof and is released under the MIT license. It is supported on Linux, MS Windows and MacOS. Additional documentation is available at the package website (https://keyes-timothy.github.io/tidytof/).

Supplementary information

Supplementary data are available at Bioinformatics Advances online.

Technology meets TILs: Deciphering T cell function in the -omics era

T cells are at the center of cancer immunology because of their ability to recognize mutations in tumor cells and directly mediate cancer cell killing. Immunotherapies to rejuvenate exhausted T cell responses have transformed the clinical management of several malignancies. In parallel, the development of novel multidimensional analysis platforms, such as single-cell RNA sequencing and high-dimensional flow cytometry, has yielded unprecedented insights into immune cell biology. This convergence has revealed substantial heterogeneity of tumor-infiltrating immune cells in single tumors, across tumor types, and among individuals with cancer. Here we discuss the opportunities and challenges of studying the complex tumor microenvironment with -omics technologies that generate vast amounts of data, highlighting the opportunities and limitations of these technologies with a particular focus on interpreting high-dimensional studies of CD8+ T cells in the tumor microenvironment.

CytofIn enables integrated analysis of public mass cytometry datasets using generalized anchors

The increasing use of mass cytometry for analyzing clinical samples offers the possibility to perform comparative analyses across public datasets. However, challenges in batch normalization and data integration limit the comparison of datasets not intended to be analyzed together. Here, we present a data integration strategy, CytofIn, using generalized anchors to integrate mass cytometry datasets from the public domain. We show that low-variance controls, such as healthy samples and stable channels, are inherently homogeneous, robust against stimulation, and can serve as generalized anchors for batch correction. Single-cell quantification comparing mass cytometry data from 989 leukemia files pre- and post normalization with CytofIn demonstrates effective batch correction while recapitulating the gold-standard bead normalization. CytofIn integration of public cancer datasets enabled the comparison of immune features across histologies and treatments. We demonstrate the ability to integrate public datasets without necessitating identical control samples or bead standards for fast and robust analysis using CytofIn.

Challenges in batch normalization and data integration limit the comparison of existing mass cytometry datasets. Here, the authors report CytofIn that can integrate mass cytometry datasets from the public domain and reveal cellular features associated with immune oncology by analyzing five public cancer datasets.

Mass Cytometry for the Characterization of Individual Cell Types in Ovarian Solid Tumors

Mass cytometry aka Cytometry by Time-Of-Flight (CyTOF) is one of several recently developed multiparametric single-cell technologies designed to address cellular heterogeneity within healthy and diseased tissue. Mass cytometry is an adaptation of flow cytometry in which antibodies are labeled with stable heavy metal isotopes and the readout is by time-of-flight mass spectrometry. With minimal spillover between channels, mass cytometry enables readouts of up to 60 parameters per single cell. Critically, mass cytometry can identify minority cell populations that are lost in bulk tissue analysis. Mass cytometry has been used to great effect for the study of immune cells. We have extended its use to examine single cells within disaggregated solid tissues, specifically freshly resected tubo-ovarian high-grade serous tumors. Here we detail our protocols designed to ensure the production of high-quality single-cell datasets. The methodology can be modified to accommodate the study of other solid tissues.

FlowCT for the analysis of large immunophenotypic datasets and biomarker discovery in cancer immunology

Large-scale immune monitoring is becoming routinely used in clinical trials to identify determinants of treatment responsiveness, particularly to immunotherapies. Flow cytometry remains one of the most versatile and high throughput approaches for single-cell analysis; however, manual interpretation of multidimensional data poses a challenge to capture full cellular diversity and provide reproducible results. We present FlowCT, a semi-automated workspace empowered to analyze large datasets that includes pre-processing, normalization, multiple dimensionality reduction techniques, automated clustering and predictive modeling tools. As a proof of concept, we used FlowCT to compare the T cell compartment in bone marrow (BM) vs peripheral blood (PB) of patients with smoldering multiple myeloma (MM); identify minimally-invasive immune biomarkers of progression from smoldering to active MM; define prognostic T cell subsets in the BM of patients with active MM after treatment intensification; and assess the longitudinal effect of maintenance therapy in BM T cells. A total of 354 samples were analyzed and immune signatures predictive of malignant transformation in 150 smoldering MM patients (hazard ratio [HR]: 1.7; P <.001), and of progression-free (HR: 4.09; P <.0001) and overall survival (HR: 3.12; P =.047) in 100 active MM patients, were identified. New data also emerged about stem cell memory T cells, the concordance between immune profiles in BM vs PB and the immunomodulatory effect of maintenance therapy. FlowCT is a new open-source computational approach that can be readily implemented by research laboratories to perform quality-control, analyze high-dimensional data, unveil cellular diversity and objectively identify biomarkers in large immune monitoring studies.

Making the most of high-dimensional cytometry data

High-dimensional cytometry represents an exciting new era of immunology research, enabling the discovery of new cells and prediction of patient responses to therapy. A plethora of analysis and visualization tools and programs are now available for both new and experienced users; however, the transition from low- to high-dimensional cytometry requires a change in the way users think about experimental design and data analysis. Data from high-dimensional cytometry experiments are often underutilized, because of both the size of the data and the number of possible combinations of markers, as well as to a lack of understanding of the processes required to generate meaningful data. In this article, we explain the concepts behind designing high-dimensional cytometry experiments and provide considerations for new and experienced users to design and carry out high-dimensional experiments to maximize quality data collection.

High Dimensional Imaging Mass Cytometry Panel to Visualize the Tumor Immune Microenvironment Contexture

The integrative analysis of tumor immune microenvironment (TiME) components, their interactions and their microanatomical distribution is mandatory to better understand tumor progression. Imaging Mass Cytometry (IMC) is a high dimensional tissue imaging system which allows the comprehensive and multiparametric in situ exploration of tumor microenvironments at a single cell level. We describe here the design of a 39-antibody IMC panel for the staining of formalin-fixed paraffin-embedded human tumor sections. We also provide an optimized staining procedure and details of the experimental workflow. This panel deciphers the nature of immune cells, their functions and their interactions with tumor cells and cancer-associated fibroblasts as well as with other TiME structural components known to be associated with tumor progression like nerve fibers and tumor extracellular matrix proteins. This panel represents a valuable innovative and powerful tool for fundamental and clinical studies that could be used for the identification of prognostic biomarkers and mechanisms of resistance to current immunotherapies.

K-means quantization for a web-based open-source flow cytometry analysis platform

Flow cytometry (FCM) is an analytic technique that is capable of detecting and recording the emission of fluorescence and light scattering of cells or particles (that are collectively called “events”) in a population¹. A typical FCM experiment can produce a large array of data making the analysis computationally intensive². Current FCM data analysis platforms (FlowJo³, etc.), while very useful, do not allow interactive data processing online due to the data size limitations. Here we report a more effective way to analyze FCM data on the web. Freecyto is a free and intuitive Python-flask-based web application that uses a weighted k-means clustering algorithm to facilitate the interactive analysis of flow cytometry data. A key limitation of web browsers is their inability to interactively display large amounts of data. Freecyto addresses this bottleneck through the use of the k-means algorithm to quantize the data, allowing the user to access a representative set of data points for interactive visualization of complex datasets. Moreover, Freecyto enables the interactive analyses of large complex datasets while preserving the standard FCM visualization features, such as the generation of scatterplots (dotplots), histograms, heatmaps, boxplots, as well as a SQL-based sub-population gating feature². We also show that Freecyto can be applied to the analysis of various experimental setups that frequently require the use of FCM. Finally, we demonstrate that the data accuracy is preserved when Freecyto is compared to conventional FCM software.

Dissecting Response to Cancer Immunotherapy by Applying Bayesian Network Analysis to Flow Cytometry Data

Cancer immunotherapy, specifically immune checkpoint blockade, has been found to be effective in the treatment of metastatic cancers. However, only a subset of patients achieve clinical responses. Elucidating pretreatment biomarkers predictive of sustained clinical response is a major research priority. Another research priority is evaluating changes in the immune system before and after treatment in responders vs. nonresponders. Our group has been studying immune networks as an accurate reflection of the global immune state. Flow cytometry (FACS, fluorescence-activated cell sorting) data characterizing immune cell panels in peripheral blood mononuclear cells (PBMC) from gastroesophageal adenocarcinoma (GEA) patients were used to analyze changes in immune networks in this setting. Here, we describe a novel computational pipeline to perform secondary analyses of FACS data using systems biology/machine learning techniques and concepts. The pipeline is centered around comparative Bayesian network analyses of immune networks and is capable of detecting strong signals that conventional methods (such as FlowJo manual gating) might miss. Future studies are planned to validate and follow up the immune biomarkers (and combinations/interactions thereof) associated with clinical responses identified with this computational pipeline.

Single-Cell Immune Profiling in Coronary Artery Disease: The Role of State-of-the-Art Immunophenotyping With Mass Cytometry in the Diagnosis of Atherosclerosis

Coronary artery disease remains the leading cause of death globally and is a major burden to every health system in the world. There have been significant improvements in risk modification, treatments, and mortality; however, our ability to detect asymptomatic disease for early intervention remains limited. Recent discoveries regarding the inflammatory nature of atherosclerosis have prompted investigation into new methods of diagnosis and treatment of coronary artery disease. This article reviews some of the highlights of the important developments in cardioimmunology and summarizes the clinical evidence linking the immune system and atherosclerosis. It provides an overview of the major serological biomarkers that have been associated with atherosclerosis, noting the limitations of these markers attributable to low specificity, and then contrasts these serological markers with the circulating immune cell subtypes that have been found to be altered in coronary artery disease. This review then outlines the technique of mass cytometry and its ability to provide high-dimensional single-cell data and explores how this high-resolution quantification of specific immune cell subpopulations may assist in the diagnosis of early atherosclerosis in combination with other complimentary techniques such as single-cell RNA sequencing. We propose that this improved specificity has the potential to transform the detection of coronary artery disease in its early phases, facilitating targeted preventative approaches in the precision medicine era.

Stem cell imaging through convolutional neural networks: current issues and future directions in artificial intelligence technology

Stem cells are primitive and precursor cells with the potential to reproduce into diverse mature and functional cell types in the body throughout the developmental stages of life. Their remarkable potential has led to numerous medical discoveries and breakthroughs in science. As a result, stem cell-based therapy has emerged as a new subspecialty in medicine. One promising stem cell being investigated is the induced pluripotent stem cell (iPSC), which is obtained by genetically reprogramming mature cells to convert them into embryonic-like stem cells. These iPSCs are used to study the onset of disease, drug development, and medical therapies. However, functional studies on iPSCs involve the analysis of iPSC-derived colonies through manual identification, which is time-consuming, error-prone, and training-dependent. Thus, an automated instrument for the analysis of iPSC colonies is needed. Recently, artificial intelligence (AI) has emerged as a novel technology to tackle this challenge. In particular, deep learning, a subfield of AI, offers an automated platform for analyzing iPSC colonies and other colony-forming stem cells. Deep learning rectifies data features using a convolutional neural network (CNN), a type of multi-layered neural network that can play an innovative role in image recognition. CNNs are able to distinguish cells with high accuracy based on morphologic and textural changes. Therefore, CNNs have the potential to create a future field of deep learning tasks aimed at solving various challenges in stem cell studies. This review discusses the progress and future of CNNs in stem cell imaging for therapy and research.