Hello and welcome guinea pigs/ lab friends!

let us demo some genome view

a quick disclaimer: this is unfinished software! Things may be funky/broken/unintuitive. Thanks in advance for your time, patience, and infinitely helpful feedback.

Let's get the genome view development branch!

From your local anvi'o codebase, run git checkout genomeview-backend to pull + sync the remote genomeview-backend branch.

Genome View relies on an additional submodule called Fabric.js . We can make sure we have that by running git submodule update --init

You should now be able to run anvi-self-test --suite genome-view without any hiccups. Open the resultant URL in Chrome, and lets go!

Getting situated

view The Genome View interface is comprised of three main areas

  1. The 'main canvas' - this is where your visualization lives
  2. The 'nav bar' - your one stop shop for quickly navigating the main canvas
  3. The 'settings pane' - an anvi'o classic! Easily accessible with the s shortcut

Navigating the main canvas

Manually navigating the main canvas can happen in a few different ways.

  • Mousewheel up/down over the main canvas will zoom in/out
  • Manipulating the nav bar (slide, condense, expand) will adjust the main canvas accordingly
  • Supplying start/stop values for the input fields on the bottom left of the nav bar, followed by enter
  • Clicking the + and - buttons on the bottom right of the nav bar will zoom in/out
  • The main canvas also has a verticle scrollbar of its own to view obscured genome groups

Learning about our genes

Genome View offers two main avenues for gene-level information

  • The 'ephemeral tooltip', available by hovering over a gene arrow
  • The 'deepdive tooltip', available by clicking on a gene arrow

Go ahead and mouse over any gene in the main canvas

  • Is all the expected information rendered in the tooltip?
  • Is the rendered data laid out sensibly?

Now, click on any gene to open the deepdive tooltip

  • Is it immediately clear what additional information/UI is exposed here?
  • Try blasting a gene - is this feature functioning as expected?

While we have the deepdive tooltip open, let's explore metadata

  • Metadata is divided into two categories
    • tags - for short, keyword descriptors
    • descriptions - for longer, more thorough text associated with a given gene.
  • Add a test metadata tag to a handful of arbitrary genes. I like to use 'cats' :)
  • Your tags should immediately populate the metadata section of the tooltip
  • You can query for tags against the entire sequence and instantly see your hits
    • In the metadata section, click "Query Sequence for Matches" for a given tag
    • Your visualization should:
      • zoom out to a depth containing all hits
      • glow each matched gene in the visualization for 5 seconds*
        • *the glow effect is slightly broken right now ;)
  • You can also query metadata from the settings pane. Speaking of...

Exploring the settings pane

*At present, the settings pane contains a number of placeholder/in-production elements

Here's a brief overview of the various sub-sections in the settings pane, in order of appearance.

  • Genome Arrangement - where you can change the genome spacing (vertical height) value of each genome group, as well as arrange groups by user-supplied ordering methods. Gene-cluster functionality is a work in progress :)
  • Search - a versatile query tool
  • Genomes - mostly placeholder data here, although genomes can be clicked-and-dragged to rearrange them in the visualization
  • Group Layers - rearrange the constituent layers for each group. Show/hide layers, and recolor additional data layers
  • Scale - change values for the ruler layer in each genome group
  • Bookmarks - bookmarks are metadata at the sequence level. A bookmark saves the start/stop positions of your current zoom-level along with a title and description. Bookmarks will continue to be built out to persist additional session data.
  • Gene Arrows/Links/Labels - all kinds of gene visualization stuff! too much to list!

Try to do some things!

  • Pick any gene and choose an annotation item to query - then try querying from the settings pane
  • Did the above query zoom to a specific section of the sequence? Try saving a bookmark that reflects your search criteria.
  • Change the "Color genes by" value from default to COG_CATEGORY
    • Check the "Link gene/label source" checkbox and select a new "Color genes by" value
    • Batch color gene-arrows by any criteria you'd like. Does this work as expected?
    • Reset function colors (bug* move the navbar to trigger a re-render)
  • We can also edit cosmetics of multiple genes at once manually using the lasso menu
    • Use the mouse drag box to select multiple genes from a single genome
    • From this menu, batch color the selected genes and add a gene set label which will appear above the individual gene labels
    • Now, make an additional selection that includes some of the genes you selected previously. You should notice that you are not able to create a set - label for these genes since they overlap with an existing label

Some closing thoughts

  • What worked as expected? what didn't?
  • Was there any core functionality that you felt was lacking? missing entirely?