Instructor Notes
General Notes
This lesson is designed to run in roughly half a day (~1.5 hours teaching, ~2.25 hours of hands-on exercises, plus breaks). It assumes learners already have IGV installed and the workshop data downloaded (see Setup) — do not spend workshop time on installation unless you have build in extra time for it.
Episodes 1–2 need an internet connection (they load example data from the IGV server). Episodes 3–5, 7, and 8 use only local files. Episode 6 needs an internet connection again for the hosted RNA-seq tutorial data. Episode 9 is mostly discussion/demonstration based, walking through published screenshots, but its hands-on section streams two public files from NCBI over the internet (a ~13 GB CRAM and a ~175 GB BAM, the latter with a 52 MB index) — check this works from the workshop network in advance, since some institutional networks block or throttle large range requests to external FTP/HTTPS servers, and the larger index in particular may take a noticeably longer time to load.
Pacing advice
It is fine not to get through every episode. Viewing Long-Read Sequencing Data is explicitly optional — cover it only if time allows, since it does not use the workshop data and is not needed for any later episode. If time is short beyond that, the episodes most worth cutting (in order) are:
- Loading Your Own Genome — can be skipped if learners will only ever use IGV’s built-in hosted genomes.
- Precomputing Coverage with igvtools — a useful but non-essential workflow detail.
The core episodes — IGV Basics, Viewing Alignments and SNPs, and Viewing Variants and Genotypes — cover the skills learners are most likely to need immediately afterward.
Common sticking points
- Loading the wrong thing from “Load from Server.” Learners sometimes check the box next to the Tutorials folder itself instead of opening it and checking a specific dataset inside, which loads far more tracks than intended. Watch for this during the first exercise and correct it early — it is much easier to prevent than to explain a cluttered screen full of unexpected tracks afterward.
-
Loading the
.baifile. Learners occasionally try to load the BAM index file directly. Remind them IGV finds it automatically. - Mixing up the Genomes menu and the File menu. Loading a genome sequence uses Genomes > Load Genome from File…; loading every other kind of track uses File > Load from File…. This distinction trips people up in Loading Your Own Genome.
- Not zooming in far enough before trying to sort/color reads by base in Viewing Alignments and SNPs — the sort/color options operate on whatever is currently centered in the view, so learners need to be at a high enough zoom level that the mismatched base is clearly visible and centered first.
- Forgetting “View as pairs” in Viewing Structural Variants — without it, mates are drawn as independent reads and the discordant orientation signature at the inversion breakpoint is much harder to spot.
Adapting this lesson
The exercises are built around a small set of example files (see Setup) chosen because they are small enough to distribute easily and show clean, unambiguous examples of each concept (a likely-real SNP vs. a likely-artifact SNP, a clear case of alternative splicing, a population-stratified variant). If you have your own lab’s data that shows similar patterns, consider substituting it to increase learner engagement, but keep a copy of the original example data as a fallback in case your own data doesn’t display as cleanly at the zoom levels used in the exercises.