Batch script does not include junctions in exported image

[Binlop]

I used this script in python to get screenshots with pre-created coordinates of genomic inserts. Tracks with readings and coverage are displayed, but the track with junctions is missing. The file size is about 6.5 GB, and the average insertion length is about 2-6 thousand nucleotides. IGV version 2.19.7. What could be the problem?

#!/usr/bin/env python3
import subprocess

def parse_coordinates_file(filename):
    coordinates = []
    with open(filename, 'r') as f:
        for line in f:
            parts = line.strip().split('\t')
            if len(parts) >= 2:
                coordinates.append((parts[0], parts[1]))
    return coordinates

def create_igv_batch_script(coordinates, output_dir, bam_files=None, custom_files=None, genome="hg38"):
    batch_commands = []
    batch_commands.extend([
        f"genome {genome}",
        "preference SQUISH_JUNCTION_TRACK=false",   
        "preference JUNCTION_TRACK_MIN_READS=1",
        "preference SHOW_JUNCTION_TRACK=true"
    ])

    if bam_files:
        for bam_file in bam_files:
            batch_commands.extend([
                f"load {bam_file}",
            ])

    for custom_file in custom_files:
        batch_commands.append(f"load {custom_file}")

    for gene, coord in coordinates:
        safe_name = f"{gene}_{coord.replace(':', '_').replace('-', '_')}"
        batch_commands.extend([
            f"goto {coord}",
            f"snapshot {output_dir}/{safe_name}.png"
        ])
    
    batch_commands.append("exit")
    return "\n".join(batch_commands)

def main():
    input_file = "../LINE/reverse_annotation_LINE/inserts.txt"
    output_dir = "/home/nock/VSC_Projects/LINE_expression/src/igv_screenshots"
    
    custom_files = [
        "/home/nock/VSC_Projects/LINE_expression/LINE/L1_igv.sorted.bed"
    ]
    
    coordinates = parse_coordinates_file(input_file)
    batch_script = create_igv_batch_script(
        coordinates=coordinates,
        output_dir=output_dir,
        custom_files=custom_files,
        bam_files=['/home/nock/NIMC_RNA_Seq/filtered_bam/Placenta504_sorted.bam']
    )
    
    with open("igv_batch.txt", 'w') as f:
        f.write(batch_script)
    
    subprocess.run(['igv', '-b', 'igv_batch.txt'])
    # os.remove("igv_batch.txt")

if __name__ == "__main__":
    main()

[jrobinso]

I see the junctions track in the screenshot, however no junctions. What does it look like from the UI (i.e. not using a batch script)?

[Binlop]

If you just look at it from the UI, then junctions are available

I fixed the inconsistency in the form of the sample name in the UI and in the batch. I ran batch mode on the 520 sample as well. And I also got the lack of junctions

At the same time, if I use the names of the genes in batch mode, junctions are visible in the screenshots. For example: goto KIF7. All settings for displaying junctions in IGV are used by default.

[jrobinso]

I don't think its gene names per se, its more likely the size of the region queried. But that is just a guess. Your first screenshot is very zoomed in, neither side of the junctions are in view. As a workaround or test try this, first go to a zoomed out region, say 10kb, centered on the region you want a screenshot in. Then zoom into the region you want and take the screenshot. The data from the zoomed out view should still be available in the local cache. Do you see junctions in that case?

[Binlop]

I added a preload of +- 40,000 nucleotides at once. For example, the same KIF7 gene has a length of about 40,000 thousand nucleotides. However, my junctions did not load.

  for gene, coord in coordinates:
      chr = coord.split(':')[0]
      start = int(coord.split(':')[1].split('-')[0])
      end = int(coord.split(':')[1].split('-')[1])
      safe_name = f"{gene}_{coord.replace(':', '_').replace('-', '_')}"
      batch_commands.extend([
          f"goto {chr}:{start-20000}-{end+20000}",
          f"snapshot {output_dir}/{safe_name}_large.png",
          f"goto {coord}",
          f"snapshot {output_dir}/{safe_name}.png"
      ])

[jrobinso]

So far I can't reproduce this issue. Could you try the following and let us know what you see?

new
genome hg38
load https://downloads.pacbcloud.com/public/dataset/Kinnex-full-length-RNA/DATA-Revio-IDT-BrainUHRR/3-FLNC-mapped/flnc-bc01-brain-IDT-rep1.mapped.bam
goto chr12:98,593,331-98,602,812
snapshot longread_rna.png