guigolab / bamstats Goto Github PK

The following compression format would be useful:

• gzip
• bzip2

• remove BedTools dependency
• reads gene and exon features and generates the genomic elements for the index

Hello,

When trying to run jobs with the GUI I get the error "-t requires an argument".

I have tried specifying a target through the following example of a job:

java -Xmx6g -jar BAMStats-1.25/./BAMStats-GUI-1.25.jar -i X.bam -t ref.chr.bed -o X.out

I have generated the target file as follows:

samtools faidx genome.fasta
cut -f1,2 genome.fasta.fai > genome.size.tab

or

awk -F'\t' '{print $1"\t0\t"$2}' ../genome.fasta.fai > ref.chr.bed

Any help is appreciated!

Stacktrace:

INFO[0000] Running bamstats 0.3.2-dev
INFO[0000] Using 8 out of 8 logical CPUs
INFO[0000] Creating index for gencode.v29lift37.annotation.nochr.gtf.gz
INFO[0005] Annotation scanned: 36 chromosomes found
panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x1 addr=0x20 pc=0x54e3cf]

goroutine 132 [running]:
github.com/guigolab/bamstats/annotation.(*Feature).Element(...)
        /home/emilio/workspace/bamstats/annotation/feature.go:96
github.com/guigolab/bamstats/annotation.updateIndex(0xc044112000, 0x0, 0x41054fe000000000, 0x630ddc, 0x4, 0x631f26, 0xa, 0x1, 0xc000084180, 0x0)
        /home/emilio/workspace/bamstats/annotation/annotation.go:192 +0x47f
github.com/guigolab/bamstats/annotation.createTree(0xc000084120, 0xc043b00ac0, 0xa, 0x41054fe000000000, 0xc055df4300, 0xc000019500, 0xc000084180)
        /home/emilio/workspace/bamstats/annotation/annotation.go:215 +0x12a
created by github.com/guigolab/bamstats/annotation.createIndex
        /home/emilio/workspace/bamstats/annotation/annotation.go:255 +0x268

https://github.com/IHEC/ihec-assay-standards/tree/199ec96/qc_metrics/rna-seq

see this comment from golang/go#13809

Using the packaged bamstats-v0.3.5-linux-x86_64 or a locally compiled (go1.15.6 linux/amd64) dev version on CentOS 7, bamstats will crash with "panic: <fileID.bam> too many open files" if the number of BAM references contained therein exceeds the system file open limit, given by ulimit -n (often 1024 but can vary considerably among machines).

For many high-quality genomes, the number of chromosomes, scaffolds and contigs is generally much less than 1024 - no problem for bamstats. However, in many draft genomes, this is no longer true thus limiting the applicability of bamstats to a broader audience, at least those who work in draft genome assemblies.

I have never worked in go but I suspect that this feature may result from not closing streams that were opened implicitly. In any case, I hope you can address this issue.

Ensembl GTFs have "gene_biotype", instead of "gene_type", attribute.
ftp://ftp.ensembl.org/pub/release-102/gtf/homo_sapiens.

This affects the calculation of fraction_rrna in the bamstats program, if nothing else.

Hello,

Thank you for the tool.

Unfortunately, I am having trouble using the tool. I clone the repo but I am not able to run it. Can someone please provide examples of how can I use the tool?

Thanks.

Best,
Tushar

It depends on the implementation. Need to infer the type correctly. At the moment uint8 is expected and a panic is thrown if it is different (e.g. int8):

Panic: interface conversion: interface {} is int8, not uint8

Report stats for references in the BAM file. Options for selecting the references to report could be:

• all reference in the BAM header
• reference list from a file
• reference list from a command line flag

Hi @emi80

I ran Bamststs to calculate coverage. However, I am confused between Two histograms whose axis are labeled the same. By Coverage depth which histogram represents true Vertical Depth(Coverage). What does different scale for % reference scale signify.

Check picard CollectInsertSizeMetrics options:

• DEVIATIONS
• MINIMUM_PCT

Hello,

Can you clarify how to generate the rnaseq statistics using bamstats? The readme says the following:
"Provided statistics
Bamstats can currently compute the following mapping statistics:
general
genome coverage
RNA-seq"

But when I run it on my STAR genome aligned or bowtie2 transcriptome aligned bams I only get general statistics. Is it using some information in the bam header to determine whether it is an rnaseq experiment?

Is it related to the -a parameter? I am unclear as to what that does as well.

Running the following code:

bamstats -i F137214.Aligned.out.sorted.bam -o F137214_stats.json

Thanks