Reference genome alignment

Question

[amy16]

How to specify paired end input files from different lanes for alignment using hisat2?

I get the following error: Warning: Output file 'P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L004_R1_paired.fastq,' was specified without -S. This will not work in future HISAT 2 versions. Please use -S instead. Extra parameter(s) specified: "P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L005_R1_paired.fastq,", "P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L006_R1_paired.fastq", "P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L004_R2_paired.fastq,", "P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L005_R2_paired.fastq,", "P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L006_R2_paired.fastq" Note that if <mates> files are specified using -1/-2, a <singles> file cannot also be specified. Please run bowtie separately for mates and singles. Overall time: 00:00:00 Error: Encountered internal HISAT2 exception (#1)

[amy16]

Is it ok to generate genome index using the default parameters available within the tool? or does it vary from case to case?

[GenoMax]

That should be fine. If you are using one of the model genomes then get the pre-made indexes from the link below.

[amy16]

The genome I need Cicer is not within that link.

[GenoMax]

Create your own in that case (instructions).

hisat2-build ref.fa cicer_idx

use --large-index option if your genome is bigger than 4 billion bp.

[amy16]

the predicted genome size is 740 Mbp

[amy16]

Can u please advise me if I should go for large or small genome index for Cicer genome size of 740 Mbp?

[GenoMax]

Small genome index should be fine.

Answer 1

[GenoMax]

Multiple files can be specified by using a comma separated list of the files for the -1 and -2 options.

hisat2 -f -x hg19_indexed -1 read1_R1.fq,read2_R1.fq,read3_R1.fq -2 read1_R2.fq,read2_R2.fq,read_R2.fq -S eg2.sam

in your case, something like this

hisat2 -f -x base_name_of_index -1 P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L004_R1_paired.fastq,P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L005_R1_paired.fastq,P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L006_R1_paired.fastq -2 P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L004_R2_paired.fastq,P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L005_R2_paired.fastq,P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L006_R2_paired.fastq -S P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT.sam

[amy16]

Do I just specify the file name or do I have specify the path as well?

I get this error: /tmp/slurmd/job4228023/slurm_script: line 17: P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L003_R1_paired.fastq,P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L004_R1_paired.fastq,P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L005_R1_paired.fastq,P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L006_R1_paired.fastq: No such file or directory

Although it has generated the output file of 15.1GB. I would like to know if this is right.. Or something has gone wrong.

[GenoMax]

You will have to use full or relative paths for all files (if they are not in the current directory). Hard to tell if the file you generated was right. You could set a variable to point to the directory your files are in e.g. MYFLS=/full_path_to_directory_w_files and then do

hisat2 -f -x base_name_of_index -1 $MYFLS/P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L004_R1_paired.fastq,$MYFLS/P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L005_R1_paired.fastq,$MYFLS/P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L006_R1_paired.fastq -2 $MYFLS/P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L004_R2_paired.fastq,$MYFLS/P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L005_R2_paired.fastq,$MYFLS/P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT_L006_R2_paired.fastq -S P1_C_1_CBBYRANXX_CGGCTATG-TATAGCCT.sam

[amy16]

Is it possible to use index file with .ebwt (bowtie2) extension for alignment using HISAT2 ?

[GenoMax]

Build your own HISAT2 indexes or download them from HISAT2 site (right hand column).