Single Cell Genome Viewer (SCGV)

SCGV is an interactive graphical tool for single-cell genomics data, with emphasis on single-cell genomics of cancer. It facilitates examination, jointly or individually, of DNA copy number profiles of cells harvested from multiple anatomic locations (sectors). In the opening view the copy-number data matrix, with columns corresponding to cells and rows to genomic locations, is represented as a heat map with color-encoded integer DNA copy number. If a phylogenetic tree is available for the cells comprising the dataset, it can be used to order the columns of the data matrix, and clones formed by closely related cells may be identified. Alternatively, the columns can be ordered by the sector of origin of the cells. Cyto-pathological information may be displayed in a separate view, including sector-specific slide images and pathology reports. Genomic sub-regions and random subsets of cells can be selected and zoomed into. Individual or multiple copy-number profiles may be plotted as copy number against the genomic coordinate, and these plots may again be zoomed into. Chromosomal regions selected within the profiles may be followed to UCSC genome browser to examine the genomic context.

Short video introduction to SCGV can be found here:

Anaconda Environment Setup

Install Anaconda

• Go to anaconda web site https://www.continuum.io/downloads and download the latest anaconda installer for your operating system. SCGV uses Python 3.6 or later so you need to choose the appropriate Anaconda installer.

• Install anaconda into suitable place on your local machine following instructions from https://docs.continuum.io/anaconda/install

Create Anaconda scgv environment

• To create scgv environment use:

  conda create -n scgv

• Activate the newly created environment with:

  conda activate scgv

Conda installer for SCGV

• To install SCGV you can use KrasnitzLab Anaconda channel:

  conda install -c krasnitzlab scgv

• To run the SCGV viewer run following command:

  scgview

Dataset Directory Structure

• Files in the dataset should conform to the following naming convention. Each filename should end with two dot-separated words. The last word is the usual file extension and second to last is the file type. For example:

  example.featuremat.txt

  is a txt file, that contains featuremat used by the viewer.

• Example dataset directory is located in subdirectory exampledata/example.directory of the project main directory. The content of the example dataset directory is as follows:

  .
  ├── example.cells.csv
  ├── example.clone.txt
  ├── example.genome.txt
  ├── example.guide.txt
  ├── example.featuremat.txt
  ├── example.features.txt
  ├── example.ratio.txt
  ├── example.seg.txt
  ├── example.tree.txt
  └── pathology
      ├── 9727420.color.map.060414.png
      ├── Area1.Benign.jpg
      ├── Area1.Benign.txt
      ├── Area2.PIN.with.Benign.jpg
      ├── Area2.PIN.with.Benign.txt
      ├── Area3.GS9.invading.SV.jpg
      ├── Area3.GS9.invading.SV.txt
      ├── Area4.GS9.near.Urethra.jpg
      ├── Area4.GS9.near.Urethra.txt
      ├── Area5.GS9.at.Capsule.jpg
      ├── Area5.GS9.at.Capsule.txt
      └── description.csv

• Optionally the dataset directory may contain a pathology subdirectory that contains pathology images and notes. This subdirectory should contain a file called description.csv with the following structure:

  sector,pathology,image,notes
  1,Benign prostatic tissue,Area1.Benign.jpg,Area1.Benign.txt
  2,Pin and benign prostate,Area2.PIN.with.Benign.jpg,Area2.PIN.with.Benign.txt
  3,Gleason 9 and invading seminal vesicle,Area3.GS9.invading.SV.jpg,Area3.GS9.invading.SV.txt
  4,Gleason 9 near urethra,Area4.GS9.near.Urethra.jpg,Area4.GS9.near.Urethra.txt
  5,Gleason 9 at capsule,Area5.GS9.at.Capsule.jpg,Area5.GS9.at.Capsule.txt
  

First column in description.csv contains the name/id of the sector as specified in the guide file, the second column is a description of the sector and the last two columns contain file names of pathology image and notes.

Dataset Archive Structure

• Viewer supports datasets stored as a ZIP archives. Files from the archive dataset should follow the same naming convention as for dataset directories.

• Example dataset ZIP archive is found in exampledata/example.archive.zip into project main directory. The structure of the example dataset is as follows:

  unzip -t example.archive.zip 
  Archive:  example.archive.zip
      testing: pathology/               OK
      testing: pathology/Area5.GS9.at.Capsule.jpg   OK
      testing: pathology/Area4.GS9.near.Urethra.jpg   OK
      testing: pathology/Area3.GS9.invading.SV.jpg   OK
      testing: pathology/Area2.PIN.with.Benign.jpg   OK
      testing: pathology/Area1.Benign.jpg   OK
      testing: pathology/Area5.GS9.at.Capsule.txt   OK
      testing: pathology/Area4.GS9.near.Urethra.txt   OK
      testing: pathology/Area1.Benign.txt   OK
      testing: pathology/description.csv   OK
      testing: pathology/Area2.PIN.with.Benign.txt   OK
      testing: pathology/Area3.GS9.invading.SV.txt   OK
      testing: example.cells.csv        OK
      testing: example.clone.txt        OK
      testing: example.featuremat.txt   OK
      testing: example.features.txt     OK
      testing: example.genome.txt       OK
      testing: example.guide.txt        OK
      testing: example.ratio.txt        OK
      testing: example.seg.txt          OK
      testing: example.tree.txt         OK
  No errors detected in compressed data of example.archive.zip.

Start the Viewer

• Before starting the viewer you need to activate viewer's Anaconda environment

  conda activate scgv

• To start the viewer use:

  scgview

• The startup window of SCGV should appear:

1-open directory; 2-open archive; 3-open feature view; 4-additional tracks confguration; 5-list of selected profiles; 6-show selected profiles; 7-clears list of selected profiles; 8-heatmap legend; 9-sectors legend; 10-order view by sectors; 11-track legend selector; 12-selected track legend; 13-order view by selected track

Select Dataset

• Use Open Directory (1) and Open Archive (2) buttons to open a data set for visualization

• Open Directory button allows you to select a directory where a dataset is located. One directory may contain only one dataset.

• Open Archive button allow you to select dataset stored as a ZIP archive.

Viewer Main Window

• After dataset is loaded it will displayed into the main window.

• From profiles instruments you can select individual cells to display their CN profile into single profile viewer.

• Buttons 'Feature View' and 'Order by Sector View' will display different views of the whole dataset

• From 'Sectors' legend you can visualize single sector view and pathology view for any given sector.

Copy-number Profile Tools

• If you right click on a single cell it will be added to list of profiles to visualize from 'Show Profiles' button.

• To show the selected profiles you need to click on 'Profiles Show' button. Selected profiles will be visualized as stacked plot of copy number against the genomic coordinate:

1-reset view; 2-undo; 3-redo; 4-move the image; 5-zoom in; 6-save an image snapshot; 7,8-individual profiles stacked view

• Using zoom button (5) and reset button (1) you can inspect different regions of the profile.

• To examine the genomic content of an intra-chromosomal region, right-click and use the context menu to select desired genomic position or region in the stacked copy-number profile view. This will invoke UCSC Genome Browser in a tab of your default web browser

Sectors Legend

• Right click on a sector in Sector legeng will open a context menu:

  

• Selecting 'Show sector view' displays single sector view for the selected sector

  

• Selecting 'Show sector patology' will display pathology image and notes for the selected sector

  

Configurable Tracks

Using 'Configure Tracks' (6) button you can configure tracks to be shown on bottom of the heatmap view.

• When you click 'Configure Tracks' (6) button a dialog window is shown allowing you to select additional tracks to be displyed

  

• Once selected the configured tracks are shown at the bottom of the heatmap view and are available from 'Track legend selector' (11):

  

• The 'Selected track legend' (12) allows you additionally explore the selected track using context menu: