two questions in loading data

Question

[hankblues.han]

hello,

I am new in using the IGB, now i have two questions.

First,I want to visualize the ChIP-seq data through IGB .its ok to load the genome and bed file, but when i open the bam file, it always note me "Region in view is big(>500k),do you want to continue?" .after click ok ,there is still no data for the bam file. what should I do to visualize the bam file ??

Second, when I want to load the genome sequence, it always fails, but my internet is good and the speed is ok, why I cannot load the sequence successfully?

Answer 1

[dcnorris]

Hi hankblues.han,

I am the lead developer on the IGB project and wanted to offer some advice as well as offer to help you troubleshoot the issue with loading your data. First, I strongly advise you to use the latest version of IGB (no longer provided through webstart). We have not supported java webstart for over a year, so you will be running a very old version of our software if you choose to use webstart. To download the latest version of our software please visit http://bioviz.org/igb/download.html. On this page, we provide native installers for each OS (Linux, Windows, and OSX) with a bundled Java Runtime version. By default the bundled java runtime we include will consume ~25% of available system memory (i.e. Ram) for the IGB process. If this is not enough, you can configure the amount of memory (see https://wiki.transvar.org/display/igbman/Get+IGB#GetIGB-HowtochangeIGBmemoryusage)

I think watching some of our educational videos on using IGB may also help you eliminate the memory consumption issues (please see our youtube channel at https://www.youtube.com/channel/UC0DA2d3YdbQ55ljkRKHRBkg).

I specifically would recommend the following videos:

RNA-Seq Alignment and Visualization (Focus on a Feature)

ChIP-Seq Analysis and Visualization

If you continue to have any issues, I am happy to setup a video conference call to assist.

Best Regards,
David Norris

Answer 2

[michael.ante]

Hi,

usually I work with IGV, but there are sometimes similar problems:

• How big is your RAM / how much RAM have you allocated for IGB? Having lots of reads to display IGV consumes a lot of memory.
• Have you indexed your bam-files?
• Are the names of the chromosomes the same in your alignments and in the IGB-annotation? (The UCSC/Refseq annotations start with a 'chr' whilst e.g. ENSEMBL leaves this out)

To circumvent the large file issues, you can transform your bam-file into a bigwig file or a bedgraph.

Can you download the genome sequence and load it locally into your IGB?

Cheers,
Michael

[hankblues.han]

hi, Michael,

thanks a lot for your reply.

1. my bam file is about 200M ,I didn't allocate the RAM for my IGB, how to allocate the RAM? Besides, my computer has a RAM for 32G, should I allocate 8G(1/4) for IGB software?
2. yes I indexed myu bam file
3. they are the same cause my annotation also start with 'chr'
4. I also want to download the genome sequence to my local computer, but I have no idea how to download it, just download it from UCSC? which file format is ok, FASTA?

yours,
Hank

[michael.ante]

Hi Hank,

Do you run the IGB as webstart or locally? In case of webstart use as much RAM as possible, otherwise run it from the commandline as suggested here and provide 8 to 10 G.

You can download the annotation here just browse to your genome. Usually, it's given in FASTA format. Eventually, each chromosome is in a different file. You should concatenate them in the same order as in given order in the bam-header (samtools view -H my.bam).

[hankblues.han]

Hi,Michael

I run the IGB as webstart, got it and thank you very much!

Answer 3

[Vivek]

To add to what Michael said, unless you are looking for CNVs that are large, you can use samtools to subset your BAM file and load smaller files into the viewer or try an alternative like IGV to see if it makes a difference.

[hankblues.han]

ok ,i will try ,thank you