[GenoMax]

Something does not appear to be right (are you using the correct genome).

531380 (0.86%) aligned exactly 1 time
16490453 (26.84%) aligned >1 times
27.70% overall alignment rate

Are you using new version of *ALL* programs (samtools/bcftools) in this pipeline?

BTW: Please create new threads for new questions. Your question is not related to the original thread.

[zillur]

Thank you very for your kind reply. I have installed all the smalls/bcftools pipeline approximately 3 months ago. I think at that time these were latest. I have no idea about any update.

Best Regards
Zillur

[kaps]

How can I install and run bowtie on a 32 bit windows computer?

[GenoMax]

How can I install and run bowtie on a 32 bit windows computer?

You can't directly install/run bowtie2 on windows. You will have to use a unix like environment (cygwin) or a virtual machine running linux. Both options would encounter limits on a 32-bit machine because of amount of memory programs can use (how much RAM do you have?) and the size of NGS datasets.

Do you have access to a unix server (through your university/company)? It may be much better to do this type of analysis there.

[kaps]

Installed RAM IS 2 GB.
I have access to a unix server.
There are a number of bowtie modules installed already( the latest is bowtie2/2.2.5). Trying to read through the Bowtie manual so that I try analysis on the server. A quick guide from you or any other member on the forum would quickly start me off!

[GenoMax]

Installed RAM IS 2 GB.
I have access to a unix server.
There are a number of bowtie modules installed already( the latest is bowtie2/2.2.5). Trying to read through the Bowtie manual so that I try analysis on the server. A quick guide from you or any other member on the forum would quickly start me off!

Use the server for all your work. With just 2G of RAM the local computer should only be used to access the server.

I am going to point you to a guide about how to use bowtie2: http://homer.salk.edu/homer/basicTutorial/mapping.html.

You will need a multi-fasta formatted file for your virus sequences (blast db files are not usable).

Broad steps:

1. Create an index for your virus sequences (bowtie2-build). This is what you are going to search against.

$ /path_to/bowtie2-build virus.fa VIR

VIR would become the "basename" for the bowtie2 index.

2. Start with trimmed/qc'ed fastq data.
3. Do the alignments (bowtie2) and save results in sam format file.

Single-end reads

$ /path_to/bowtie2 -p 2 -x VIR -U sample1_file.fastq -S sample1.sam

Paired-end reads

$ /path_to/bowtie2 -p 2 -x VIR -1 sample1_R1_file.fastq -2 sample1_R2_file.fastq -S sample1.sam

-p 2 is number of cores used for search. Keep to <= 4 for now. Provide "basename" (VIR) for index for -x option. Do not include file extensions.

4. Use samtools to "view/sort/index" sam file to bam format.
5. Evaluate the bam file.

[kaps]

Tried the alignment but I am getting an error:

here is the script: bowtie2 -p 2 -x RNAseq4 -v 4_S4_L001_R1_001_paired_trimmed_paired.fa RNAseq4.sam
and here is report on the terminal:

Command: /export/apps/bowtie2/2.2.5/bowtie2-align-s --wrapper basic-0 -p 2 -x RNAseq4 -v 4_S4_L001_R1_001_paired_trimmed_paired.fa RNAseq4.sam
(ERR): bowtie2-align exited with value 1

where am going wrong?

[GenoMax]

You have a "v" instead of a "U" before your read file (if the command above is correct).

If you have paired-end reads (which it seems you do, from the R1 in your file name) you should map the pairs together at the same time using the example command I have included above.

[kaps]

Hi GenoMax,

I rectified the typing error (U instead of v). however, when i try to run bowtie, this the comment I get: bowtie2 -x RNAseq4 -U 4_S4_L001_R1_001_paired_trimmed_paired.fa -S RNAseq4.sam
Error: reads file does not look like a FASTQ file
terminate called after throwing an instance of 'int'
(ERR): bowtie2-align died with signal 6 (ABRT)

I imported the read file (.fa) from Clc genomics workbench (The software pairs the forward-R1 and reverse-R2 into one file).

I have decided to use the raw reads (R1 and R2) as obtained from illumina. I have however only trimmed with out first performing a quality filter. and i have proceeded with bowtie for paired end reads. Will these affect the alignment results?

[GenoMax]

You should use fastq reads for the bowtie2 alignment (you are back on the right track). Unless you know that you have data with Q scores less than 10 (e.g. based on FastQC analysis) it should be fine to proceed without explicit quality score filtering.

[kaps]

Ok.

Have proceeded with the downstream tools hoping to visualize the results as below: samtools mpileup -u -f cowpea4RNAseq_ind.fa lib4seq.sorted.bam | bcftools view -cvg -> lib4seq.raw.bcf

but the response is as follows:

[fai_load] build FASTA index.
[fai_build_core] inlined empty line is not allowed in sequence 'gi|325930233|gb|JF427592.1|'.
[fai_load] fail to open FASTA index.
-bash: bcftools: command not found

Not sure where I am going wrong!

[GenoMax]

If you want to visualize the alignments then use Integrated Genomics Viewer: http://www.broadinstitute.org/igv/ You have to sort and index your BAM file (which you seem to have done). Cowpea genome is not going to be included IGV so you will need to supply a fasta (make a note) formatted file of the genome sequence to IGV. Use the same file you used for making the bowtie2 indexes.

As for the error above "bcftools" is a program from samtools package. It is included in the source and needs to be compiled (if your administrators have not done so ask them to do that). Only required if you are going to do SNP calling.

[kaps]

Any additional clues on how to run igv (igv or igvtools?)?

[GenoMax]

Are you having trouble using IGV? If you are running it from the server you will need to use X11/X-Windows. http://www.broadinstitute.org/igv/UserGuide

BTW: Start creating a new thread when you have new questions. The current questions are no longer related to the parent thread.

[kaps]

Use the server for all your work. With just 2G of RAM the local computer should only be used to access the server.

I am going to point you to a guide about how to use bowtie2: http://homer.salk.edu/homer/basicTutorial/mapping.html.

You will need a multi-fasta formatted file for your virus sequences (blast db files are not usable).

Broad steps:

1. Create an index for your virus sequences (bowtie2-build). This is what you are going to search against.

$ /path_to/bowtie2-build virus.fa VIR

VIR would become the "basename" for the bowtie2 index.

2. Start with trimmed/qc'ed fastq data.
3. Do the alignments (bowtie2) and save results in sam format file.

Single-end reads

$ /path_to/bowtie2 -p 2 -x VIR -U sample1_file.fastq -S sample1.sam

Paired-end reads

$ /path_to/bowtie2 -p 2 -x VIR -1 sample1_R1_file.fastq -2 sample1_R2_file.fastq -S sample1.sam

-p 2 is number of cores used for search. Keep to <= 4 for now. Provide "basename" (VIR) for index for -x option. Do not include file extensions.

4. Use samtools to "view/sort/index" sam file to bam format.
5. Evaluate the bam file.

Hello,
Why would bowtie abort with such a command?

bowtie2 -p 2 -x CABFG -U clc_trim_pair4.fq -S S4_clc_trm.sam
Error: Read No_name has more quality values than read characters.
terminate called after throwing an instance of 'int'
(ERR): bowtie2-align died with signal 6 (ABRT)