Visualizing alignments in CRISPResso2

[eul2021]

Hello,

I am using CRISPResso2 starting from paired end fastqs and wanted to visualize alignments. Should the generated alignments be somewhere in the pickle file? Is there a way to output the intermediate alignment files as BAMs so that they can be visualized in IGV for instance? I'm trying to get sort of "coverage" graphs to visualize reads mapped across entire amplicon to better understand the samples.

Thank you very much for your help.

[kclem]

Hi eul2021,

thanks for using CRISPResso2.

Generated alignments of the entire read/amplicon are in the file Alleles_frequency_table.zip.

One of the problems we have with generating bams is that the bam contains the starting position for each read with regard to the reference genome, and currently as input we only have an amplicon without genomic context. Implementing this functionality is in our pipeline, but if you'd like to take a crack at it I can give you some pointers as to where to start.

In the meantime, if you're just looking for coverage, you can get this information from the Nucleotide_frequency_table.txt file -- it's the sum of all A+C+T+G (and N depending on how you define coverage) for each column.

I hope that helps!

Kendell

[kclem]

I fixed a bug in the fastq_output that should be live in 2.2.5. Try updating to 2.2.5 and running with the flag --fastq_output. This will produce a file CRISPResso_output.fastq.gz that has metadata for every read about whether it was aligned or not. If it was not aligned, the second line will start with + ALN=NA ... and the attempted alignments and alignment scores will be given.

The pickle file (now json) just includes run information, but the sequencing reads and alignments would make this file too big.

[eul2021]

Hi Kendell, Thank you for the great news! I tried updating to 2.2.5 but it's not yet in anaconda (only 2.2.4 and 1.0.13). Should I wait or is there another way to update? Thank you.

[kclem]

Hm. The bioconda automatically updates, but it may take a day or two.

If you'd like to, you can install from source:
In your CRISPResso environment, run:

wget https://github.com/pinellolab/CRISPResso2/archive/refs/heads/master.zip
unzip master.zip
cd CRISPResso2-master/
python setup.py install

This is the way to install the latest updates without waiting for conda or docker.

[eul2021]

Hi Kendell, Thank you, yes it works except for it can't find flash. Any additional set up steps to run?
"CRISPResso -h
CRITICAL @ Wed, 22 Sep 2021 15:53:29:
You need to install and have the command #####flash##### in your PATH variable to use CRISPResso!
Please read the documentation!
"

[kclem]

If you install this into your CRISPResso2 conda environment flash should already be there.

Otherwise, if you're installing into a new env, you'll have to install flash (conda install -c bioconda flash) trimmomatic (conda install -c bioconda trimmomatic) and bowtie2 (conda install -c bioconda bowtie2).

[eul2021]

Kendell, thank you so much! Yes, the fastqc_output bug is fixed in 2.2.5 and I was able to install with your conda instructions as follows:
conda create -n crispresso225b -c bioconda flash trimmomatic bowtie2 pandas matplotlib seaborn jinja2 scipy numpy python=3

conda activate crispresso225b

cd …your_directory...

wget https://github.com/pinellolab/CRISPResso2/archive/refs/heads/master.zip
unzip master.zip
cd CRISPResso2-master/
python setup.py install
CRISPResso -h #check the version

Thank you very much for your help!:)