Deploy IGV webapp on linux server

zqfang included in Make bioinfo uncool again

2020-11-16 243 words 2 minutes

Contents

Config IGV on the server.
I have to share the inteactive results with my colleague. But I don’t like to install UCSC genomebrower in local. Instead, a light-weight one is what I need.

1. Installation

Install nodejs

if you have conda, just

1
conda install -c conda-forge nodejs

build igv-webapp

1
2
3
4
git clone https://github.com/igvteam/igv-webapp.git
cd ./igv-webapp
npm install
npm run build

2. Running the app

1
npx http-server -p 5000 path/to/igv_webapp

3. Configuration (Example)

gtf, bed, bam, bigwig … in resource directory fisrt.

prepare a json file contain all the required information of tracks. see also igv.js wiki. E.g. mm10_igv-lab.sv.json

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
import json
tracks = []
for b in ['a1.bed','a2.bed','c1.bed','c2.bed']:
    bed = os.path.basename(b)
    track =   {
        "name": bed.replace(".bed",""),
        "type": "annotation",
        "format": "bed",
        "sourceType": "file",
        "url": f"http://igv-app-03:5000/resources/{bed}", # your server name + port + file path 
        #order: Number.MAX_VALUE,
        #"visibilityWindow": 300000000,
        "displayMode": "EXPANDED"
        }
    tracks.append(track)
sv = {
"label": "IGV-Lab", # name will be shown under track's menu.
"description": "Any descriptions",
"link": "",
"tracks": tracks
}
with open('./resources/tracks/mm10_igv-lab.json', 'w') as outfile:
     json.dump(sv, outfile)

add the json to trackRegistry.json in the resource/tracks.

1
2
3
4
5
"mm10": [
    "resources/tracks/mm10_annotations.json",
    "resources/tracks/mm10_encode.json",
    "resources/tracks/mm10_igv-lab.json" # add 
]

load files

select the correct genome, e.g. mm10
In the dropdown menu of tracks, you’ll see IGV-Lab. Click it, and select the files you’ve just add.