We present pipeComp (
https://github.com/plger/pipeComp

), a flexible R framework for pipeline comparison handling interactions between analysis steps and relying on multi-level evaluation metrics. We apply it to the benchmark of single-cell RNA-sequencing analysis pipelines using simulated and real datasets with known cell identities, covering common methods of filtering, doublet detection, normalization, feature selection, denoising, dimensionality reduction, and clustering. pipeComp can easily integrate any other step, tool, or evaluation metric, allowing extensible benchmarks and easy applications to other fields, as we demonstrate through a study of the impact of removal of unwanted variation on differential expression analysis.

/pdf/pipecomp-a-general-framework-for-the-evaluation-of-21y1rqcula.pdf

pipeComp, a general framework for the evaluation of computational pipelines, reveals performant single cell RNA-seq preprocessing tools

From bench to bedside: Single-cell analysis for cancer immunotherapy

The massive growth of single-cell RNA-sequencing (scRNAseq) and methods for its analysis still lacks sufficient and up-to-date benchmarks that would guide analytical choices. Moreover, current studies are often focused on isolated steps of the process. Here, we present a flexible R framework for pipeline comparison with multi-level evaluation metrics and apply it to the benchmark of scRNAseq analysis pipelines using datasets with known cell identities. We evaluate common steps of such analyses, including filtering, doublet detection (suggesting a new R package, scDblFinder), normalization, feature selection, denoising, dimensionality reduction and clustering. On the basis of these analyses, we make a number of concrete recommendations about analysis choices. The evaluation framework, pipeComp, has been implemented so as to easily integrate any other step or tool, allowing extensible benchmarks and easy application to other fields (https://github.com/plger/pipeComp).

https://biorxiv.org/content/biorxiv/early/2020/02/02/2020.02.02.930578.full-text.pdf

pipeComp, a general framework for the evaluation of computational pipelines, reveals performant single-cell RNA-seq preprocessing tools

Simultaneous profiling of the spatial distributions of multiple biological molecules at single-cell resolution has recently been enabled by the development of highly multiplexed imaging technologies. Extracting and analyzing biologically relevant information contained in complex imaging data requires the use of a diverse set of computational tools and algorithms. Here, we report the development of a user-friendly, customizable, and interoperable workflow for processing and analyzing data generated by highly multiplexed imaging technologies. The steinbock framework supports image pre-processing, segmentation, feature extraction, and standardized data export. Each step is performed in a reproducible fashion. The imcRtools R/Bioconductor package forms the bridge between image processing and single-cell analysis by directly importing data generated by steinbock. The package further supports spatial data analysis and integrates with tools developed within the Bioconductor project. Together, the tools described in this workflow facilitate analyses of multiplexed imaging raw data at the single-cell and spatial level.

https://www.biorxiv.org/content/biorxiv/early/2021/11/13/2021.11.12.468357.full.pdf

An end-to-end workflow for multiplexed image processing and analysis

The catalog of the Drosophila immune cells was until recently limited to three major cell types, based on morphology, function and few molecular markers. Three recent single cell studies highlight the presence of several subgroups, revealing a large diversity in the molecular signature of the larval immune cells. Since these studies rely on somewhat different experimental and analytical approaches, we here compare the datasets and identify eight common, robust subgroups associated to distinct functions such as proliferation, immune response, phagocytosis or secretion. Similar comparative analyses with datasets from different stages and tissues disclose the presence of larval immune cells resembling embryonic hemocyte progenitors and the expression of specific properties in larval immune cells associated with peripheral tissues.

Toward a Consensus in the Repertoire of Hemocytes Identified in Drosophila.

With the development of single-cell RNA sequencing (scRNA-seq) technology, it has become possible to perform large-scale transcript profiling for tens of thousands of cells in a single experiment. Many analysis pipelines have been developed for data generated from different high-throughput scRNA-seq platforms, bringing a new challenge to users to choose a proper workflow that is efficient, robust and reliable for a specific sequencing platform. Moreover, as the amount of public scRNA-seq data has increased rapidly, integrated analysis of scRNA-seq data from different sources has become increasingly popular. However, it remains unclear whether such integrated analysis would be biassed if the data were processed by different upstream pipelines. In this study, we encapsulated seven existing high-throughput scRNA-seq data processing pipelines with Nextflow, a general integrative workflow management framework, and evaluated their performance in terms of running time, computational resource consumption and data analysis consistency using eight public datasets generated from five different high-throughput scRNA-seq platforms. Our work provides a useful guideline for the selection of scRNA-seq data processing pipelines based on their performance on different real datasets. In addition, these guidelines can serve as a performance evaluation framework for future developments in high-throughput scRNA-seq data processing.

/pdf/comparison-of-high-throughput-single-cell-rna-sequencing-27ebjvo3xe.pdf

Comparison of high-throughput single-cell RNA sequencing data processing pipelines.

Understanding the immune-cell abundances of cancer and other disease-related tissues has an important role in guiding cancer treatments. We propose data augmentation through in silico mixing with deep neural networks (DAISM-DNN), where highly accurate and unbiased immune-cell proportion estimation is achieved through DNN with dataset-specific training data created from partial samples from the same batch with ground truth cell proportions. We evaluated the performance of DAISM-DNN on three publicly available real-world datasets and results showed that DAISM-DNN is robust against platform-specific variations among different datasets and outperforms other existing methods by a significant margin on all the datasets evaluated.

https://biorxiv.org/content/10.1101/2020.03.26.009308v1.full.pdf

DAISM-DNN: Highly accurate cell type proportion estimation with in silico data augmentation and deep neural networks

Highly multiplexed imaging technology is a powerful tool to facilitate understanding cells composition and interaction in tumor microenvironment at subcellular resolution, which is crucial for both basic research and clinical applications. Imaging mass cytometry (IMC), a multiplex imaging method recently introduced, can measure up to 40 markers simultaneously in one tissue section by using a high-resolution laser with a mass cytometer. However, due to its high resolution and large number of channels, how to process and interpret the image data from IMC remains a key challenge for its further applications. Accurate and reliable single cell segmentation is the first and a critical step to process IMC image data. Unfortunately, existing segmentation pipelines either produce inaccurate cell segmentation results, or require manual annotation which is very time-consuming. Here, we developed Dice-XMBD, a Deep learnIng-based Cell sEgmentation algorithm for tissue multiplexed imaging data. %, which combine nuclear proteins and membrane/cytoplasm proteins as two channels of input. In comparison with other state-of-the-art cell segmentation methods currently used in IMC, Dice-XMBD generates more accurate single cell masks efficiently on IMC images produced with different nuclear, membrane and cytoplasm markers. All codes and datasets are available at https://github.com/xmuyulab/Dice-XMBD.

/pdf/dice-xmbd-deep-learning-based-cell-segmentation-for-imaging-5fdslvgev7.pdf

Dice-XMBD: Deep Learning-Based Cell Segmentation for Imaging Mass Cytometry.

Opioid-induced fragile-like regulatory T cells contribute to withdrawal

/pdf/inconsistent-prediction-capability-of-immunecells-sig-across-34ystjx7ms.pdf

Xu Xiao

Papers

Comparison of high-throughput single-cell RNA sequencing data processing pipelines.

DAISM-DNN: Highly accurate cell type proportion estimation with in silico data augmentation and deep neural networks

Dice-XMBD: Deep Learning-Based Cell Segmentation for Imaging Mass Cytometry.

Opioid-induced fragile-like regulatory T cells contribute to withdrawal

Inconsistent prediction capability of ImmuneCells.Sig across different RNA-seq datasets