best way to visualize bam file

Question

[lkianmehr]

Hello,

I want to ask you what is the best way for visualizing big file (Bam), I am going to sort bam file with samtools sort at first, then convert it to bedgraph by this command: bedtools genomecov -ibam *.bam -bg -scale 10.0, and visualize it by IGV. this way is right and certain in your idea or what else?

thanks in advance

[WouterDeCoster]

What is the aim of your visualisation? Are you specifically interested in a single gene or locus?

Answer 1

[Devon Ryan]

If you want to be able to see content regardless of the zoom level then make a bigWig files (e.g., with bamCoverage from deepTools or from the bedGraph file bedtools genomecov produces). But if you're going to zoom in and look at smaller regions (<20kb) anyway then just stick with the indexed and sorted BAM file. BAM will still be slower to render than bigWig data, but it otherwise works fine.

[lkianmehr]

the size of bam file is about 26 Gb, so I convert it to bedgraph format with Homer, and then to TDF file that would be open easy with IGV, then you mean I can't use this file to see small regions!

[Devon Ryan]

If I meant that you couldn't use that to see small regions I would have written that. TDF is an IGV-specific format, as opposed to bigWig files which can be used in other programs, thus making them generally preferable.

Answer 2

[Nicolas Rosewick]

You can direcly open your sorted and indexed bam file in IGV. It's not required to convert it to bedgraph.

[lkianmehr]

but is too big and IGV can't open it and will crush!

[h.mon]

IGV doesn't load the whole bam into the memory, only the section being active viewed. For bam files, IGV will display reads only after you zoom considerably into a small chromosomal region.

[Nicolas Rosewick]

What is the RAM of your computer ? What is the size of you bam file ? Which type of data ? If DNA-Seq (WGS, WES, ..) what will be the coverage, etc...

[e.r.zakiev]

Is it imperative to convert to bigWig in case of ChIPseq data?

If that's relevant, let's assume I'm interested in several specific loci/genes