Reference genome hg38

[hassanfa]

Workflows in balsamic are agnostic of which reference input is provided. But the correct reference still file needs to be provided. This also affects target sequencing, as all of our panel are hg19 coordinates.

• [ ] Update panels to hg38: https://github.com/Clinical-Genomics/target_capture_bed
• [x] Make sure annotation rules are provided correct genome version
• [x] Update workflow to download both reference hg19 and hg38
• [ ] Make sure cg can provide correct genome version
• [ ] Add suffix/prefix to reference.json to reflect content, Add meta/info to reference.json

How to add correct reference genome. Possible solutions:

1. Create two independent reference.json

Pro: doesn't need any changes to cli, one can make various reference.json files for different versions of genome. This can even handle other organisms! Contra: dumps bulk of work on reference generation, complexity of reference workflow will increase, not all versions of genome have same file names.

2. provide genome version as a CLI option, and BALSAMIC takes the correct version based on that

Pro: easier on user side, Contra: can't handle unconventional reference files or organisms with ease, won't be agnostic to whatever reference.json is provided, Still requires a lot of change to reference.json,

BALSAMIC version <4.2.2

Current affected rule if snakemake workflow related reference generation workflow

[hassanfa]

@keyvanelhami Have a look at this and lets discuss it on Monday.

[hassanfa]

BALSAMIC part of this is done. Rest will be done in other repositories. PRs were: #407 and #438