Visualization

The following are a list of visualization tools that enable viewing and exploration of BICCN-generated data.

Azimuth (RRID:SCR_021084) is a web application that uses annotated reference dataset to automate processing, analysis, and interpretation of new single cell RNA-seq experiment. Azimuth leverages reference-based mapping, a pipeline that inputs a counts matrix of gene expression in single cells, and performs normalization, visualization, cell annotation, and differential expression. All results can be explored within the app, and easily downloaded for additional downstream analysis.

Access Azimuth here:
‍https://azimuth.hubmapconsortium.org/

Described in paper by Hao et al. 2021.

Brain Architecture (RRID:SCR_004283). The Brain Architecture portal, active since 2012, provides a virtual microscopic brain observatory spanning multiple species, including mouse, marmoset, human and zebrafinch, and data analysis tools. The project is creating mesoscale brain connectivity maps for the mouse and the marmoset. A peta-voxel data set, with thousands of brains injected with neuronal tracers to help visualize the mesoscale connectivity, are interactively viewable. Spatial distributions of specific cell types, histological reference brains and atlases across species are also presented.

Access Brain Architecture here:
‍http://brainarchitecture.org/

Brain Cell Methylation Viewer (RRID:SCR_020954). Viewer for DNA Methylation Atlas of Mouse Brain at Single Cell Resolution. Browser to interactively explore single-cell methylome dataset including exploration of methylation diversity of one gene at single-cell or cell-type level, exploration of cell type composition of adult mouse brain dissection regions and anatomical structures, exploration of spatial distribution and methylation signature genes of one cell type.

Access Brainome Portal here:
‍http://neomorph.salk.edu/omb

Ecker Lab (https://ecker.salk.edu/), Salk Institute for Biological Studies

Described in paper by Liu et al. 2021.

Brainome Portal (RRID:SCR_018162). Genome browser viewer for single cell epigenomic and transcriptomic data (DNA methylation, ATAC-seq, RNA-seq) from mouse primary motor cortex. Search for genes to visualize genomic loci, including enhancers marked by peaks of chromatin accessibility and low DNA methylation.

Access Brainome Portal here:
‍https://brainome.ucsd.edu/annoj/BICCN_MOp/

Described in paper by Yao et al. 2021.

Mukamel Lab (https://brainome.ucsd.edu), UCSD Cognitive Science

cellxgene (RRID:SCR_021059) Created by the Chan Zuckerberg Initiative, cellxgene is a scalable, open source resource designed to make it easier for researchers without computational experience to easily share and explore single-cell datasets from various modalities in common formats. Users can visualize gene expression at the level of an entire dataset, or particular subset of cells to help identify cell types, which in turn can help identify diseases. Cellxgene enables plotting gene expression levels, or other features, against another gene to compare how these different genes are expressed across a dataset, and it integrates with existing tools like SciPy and Jupyter.

Access cellxgene here:
‍https://cellxgene.cziscience.com/

Described in paper by Megill et al. 2021.

Chan Zuckerberg Initiative (https://chanzuckerberg.com/science/programs-resources/single-cell-biology/cellxgene/)

Cytosplore Viewer (RRID:SCR_018330) is an interactive visual analysis application (Windows and MacOS) for exploration of multi-species and cross-omics single cell data published in Allen Cell Types Database and for a growing number of single cell data resources of Brain Initiative Cell Census Network. Allows interactive exploration of cellular hierarchies, transcriptome wide gene expression and metadata of individual cells, and computation of differential statistics between cell selections or clusters, and across species. Integrates linkout to the UCSC genome browser for linked cross-omics, cross-species exploration between matched gene expression, chromatin state and DNA methylation data.

Access Cytosplore Viewer here:
https://viewer.cytosplore.org/

Described in paper by Hodge et al. 2019.

Leiden Computational Biology Center (https://www.lcbc.nl/), Leiden University Medical Center

NeMO Analytics (RRID:SCR_018164) NeMO Analytics (nemoanalytics.org) is a web-based suite of data visualization and analysis tools, focusing single-cell data analysis. The portal allows users to visualize published or user-supplied multi-omic data across experiments and species side-by-side. Tools include cluster visualization based on expression/cell type, cluster comparison, projection of one dataset into another, identification of marker genes, comparison of marker genes across datasets and note taking.

Access NeMO Analytics here:
‍https://nemoanalytics.org

BICCN publications landing page: here.

NeuroMorpho.org (RRID:SCR_002145). Centrally curated inventory of digitally reconstructed neurons and glia associated with peer-reviewed publications that contains some of the most complete axonal arborizations digitally available in the community. Each cell is represented by a unique identifier, general information (metadata), the original and standardized ASCII files of the digital morphological reconstruction, and a set of morphometric features. It contains contributions from over 800 laboratories worldwide and is continuously updated as new morphological reconstructions are collected, published, and shared. Users may browse by species, brain region, cell type or lab name. Users can also download morphological reconstructions for research and analysis. Deposition and distribution of reconstruction files ultimately prevents data loss. Centralized curation and annotation aim to minimize the effort required by data owners while ensuring a unified format. It also provides a one-stop entry point for all available reconstructions, thus maximizing data visibility and impact.

Access NeuroMorpho.org here:
‍http://www.neuromorpho.org

Ascoli Lab: https://krasnow1.gmu.edu/cn3/ascoli/

Single Cell Portal (RRID:SCR_014816). The Single Cell Portal specializes in visualizing and disseminating single-cell data. The Portal allows you to use natural language and faceted search to discover other scientists’ research and share your own findings. Each study can include information on cell types, singular or multiple gene expression, and spatial transcriptomics. In addition, interactive visualizations allow you to explore cell clusters and search for related genes. If you wish to explore further, you can download a study’s data. As a free and public resource, we strive to make open science interactive and powerful.

Access Single Cell Portal here:
‍https://singlecell.broadinstitute.org/single_cell

RRID: SCR_014816

Contact: scp-support@broadinstitute.zendesk.com

Anatomy and Morphology Molecular Portal (RRID:SCR_014816). The Anatomy and Morphology Project is a joint effort with the Brain Architecture Portal and the Single Cell Portal. Our goal is to visually combine anatomic and molecular brain data in. Users can search for neuroscience data using natural language and faceted search, and then this portal retrieves molecular data housed in the Single Cell Portal and anatomical data in the Brain Architecture Portal. Each study includes information on cell types, gene expression, related genes, and spatial transcriptomics. You can also download a study’s data to explore further and build on your science.

Access Anatomy and Morphology Molecular Portal here:
‍https://singlecell.broadinstitute.org/single_cell?scpbr=biccn-anatomy-and-morphology-project

RRID: SCR_014816

Contact: scp-support@broadinstitute.zendesk.com

Transcriptomics Explorer (RRID:SCR_017567). Software tool to visualize and analyze transcriptomics data and transcriptomic cell types for mouse and human, all directly in the web browser. Explore gene expression across cell types, search for genes of interest, explore tSNE and UMAPs, and visualize a dataset’s sampling strategy to see how cells and nuclei were sampled across brain areas, cortical layer, and other dimensions. Used for Allen Cell Types Database to Browse Data: Human - Multiple Cortical Areas and Mouse - Cortex and Hippocampus, as well as other datasets.

Access Transcriptomics Explorer here:
‍https://portal.brain-map.org/atlases-and-data/rnaseq

BICCN Pipelines

Overview

The Broad Institute’s Data Sciences Platform (DSP) develops production-level data processing pipelines in collaboration with multiple consortia including BICCN. Thank you to everyone who has worked with us to create and improve these pipelines. For more details please see our BICCN collaborators and how to cite the pipelines.

Pipelines are available for multiple data types hosted in the BICCN network, including single-cell and single-nucleus transcriptomics, methylomics, and ATAC-seq data.

Pipeline	WARP WDL Code	Input Data	Overview	Terra Workspace	Publication (if applicable)
MethylC-Seq (CEMBA; RRID:SCR_021219)	CEMBA	Multiplexed single-nucleus bisulfite sequencing data	CEMBA Overview	CEMBA	Luo et al. 2017
Optimus (RRID:SCR_018908)	Optimus	10x Genomics V2 and V3 3' single-cell and single-nucleus data	Optimus Overview	Optimus
Single-Cell ATAC (scATAC; RRID:SCR_018919)	scATAC	Single-cell ATAC-seq data from nuclear isolates	scATAC Overview	scATAC	Fang et al. (2021)
Smart-seq2 Single Nucleus Multi-Sample (RRID:SCR_021312)	Smart-seq2 Single Nucleus Multi-Sample	Single-cell data generated with Smart-seq2 assays	Smart-seq2 Single Nucleue Multi-Sample Overview	Smart-seq2 Single Nucleus Multi-Sample

Pipeline Standards, Maintenance, and Availability

Pipelines are cloud-optimized and developed to ensure portability as well as data reproducibility and interoperability. To this aim, the pipelines are:

Open-access and developed with GA4GH standards.
Written in the Workflow Description Language (WDL), a community-maintained, human-readable workflow language that can run on Cromwell, a portable execution engine that can be launched anywhere, locally or in the cloud.
Containerized using public Docker instances, allowing researchers to exactly reproduce the workflow software.

The pipeline code is available from multiple sources, including GitHub, Dockstore, and the BCDC cloud computing environment (Terra).

Code is developed and maintained in the WDL Analysis Research Pipelines (WARP) repository in GitHub. Overviews and workflow code for BICCN-related pipelines are linked in the table above; additionally, relevant pipelines can be identified by typing the keyword “BICCN” in the WARP Documentation search bar.The WARP Overview details navigating the repository, pipeline development, and running the workflows.
Workflows are available for export from Dockstore, a GA4GH-compliant platform for sharing Docker-based tools. Search “warp” on Dockstore to find all WARP pipelines, including those used in the BICCN.
Workflows are also available to test on Terra, the cloud-based bioinformatics platform used for BCDC data processing. To get started, register for Terra using the registration guide. To try a pipeline, navigate to the pipeline’s workspace linked in the table above or search for the “BICCN” tag in the workspaces tag search bar. Each workspace contains downsampled data, detailed instructions for using the workflows, and cost guidelines. Learn more about Terra with the Getting Started guides.

Citing the Pipelines

Each BICCN pipeline (see table above) has a SciCrunch resource identifier (RRID) that can be cited in publications. Follow the SciCrunch citation guidelines.

Example: (Optimus Pipeline, RRID:SCR_018908)

Additionally, please refer to the table above to cite any publications associated with the pipeline.

Additional Single-cell Transcriptomic Pipeline Resources

These pipelines produce outputs and quality control metrics that can be further analyzed and visualized with downstream community resources. Tutorials for combining single-cell transcriptomic data and pipelines with common community tools are available in the following resources:

BICCN Omics Workshop Workspace

This tutorial Terra workspace is a step-by-step guide to analyzing BICCN 10x Genomics single-cell data. Using this workspace, researchers learn how to:

Import an example 10x dataset (FASTQs) from NeMO
Align example 10x FASTQs and produce a raw count matrix with quality metrics using the Optimus workflow
Filter, normalize, and cluster the raw count matrix with the Cumulus workflow
Explore single-cell data in a Seurat Jupyter Notebook

BICCN Omics Workshop Webinar Recording

This is some text insideThis Brain Initiative Cell Census Network (BICCN) virtual workshop guides you through finding data in the Neuroscience Multi-Omic (NeMO) Archive, analyzing that data in Terra through workflows (pipelines) and interactive analysis, then publishing the results to a study in the Single Cell Portal (SCP). of a div block.

BICCN Omics Workshop Blog

This high-level overview of the BICCN Omics Workshop describes the BICCN Omics workshop content and provides a link to the webinar demonstration.

Acknowledgments

We thank the following BICCN collaborators and Broad Pipelines Team members for their work on these pipelines:

MethylC-Seq (CEMBA)

Our gratitude to the Joseph Ecker Lab and special thanks to Joseph Ecker, Chongyuan Luo, Eran Mukamel, Hanqing Liu, Benjamin Carlin, Dan Moran, and Jeff Korte.

Single-Cell ATAC (scATAC)

Our gratitude to the Bing Ren Lab and special thanks to Bing Ren, Rongxin Fang, Yang Li, Sebastian Preissl, Nick Barkas, and Kylee Degatano.

Smart-seq2 Single Nucleus

Our gratitude to the Allen Institute, the Eran Mukamel Lab, and the NeMO team. Special thanks to Eran Mukamel, Fangming Xie, Zizhen Yao, Changkyu Lee, Jeff Goldy, Brian Herb, Cindy van Velthoven, Carrie McCracken, Kishori Konwar, Farzaneh Khajouei, Jessica Way, and Kylee Degatano.

Optimus

Our gratitude to Alex Dobin and the Eran Mukamel Lab. Special thanks to Kishori Konwar, Farzaneh Khajouei, Jessica Way, Ambrose Carr, Jishu Xu, Jose Soto, and Nick Barkas. This pipeline is currently being updated for the BICCN; more acknowledgments will be added as the work progresses.