About

CRISPResso is a software pipeline designed to enable rapid and intuitive interpretation of genome editing experiments. A limited web implementation is available at: http://crispresso2.pinellolab.org/ or http://crispresso.com.

Briefly, CRISPResso:

Aligns sequencing reads to a reference sequence
Quantifies insertions, mutations and deletions to determine whether a read is modified or unmodified by genome editing
Summarizes editing results in intuitive plots and datasets

Tools

CRISPResso is a suite of complementary tools:

CRISPResso - for analyzing and interpreting single experimental conditions on a single amplicon
CRISPRessoBatch - for analyzing and comparing multiple experimental conditions at the same site
CRISPRessoPooled - for analyzing multiple amplicons from a pooled amplicon sequencing experiment
CRISPRessoWGS - for analyzing specific sites in whole-genome sequencing samples
CRISPRessoCompare - for comparing editing between two samples (e.g., treated vs control)
CRISPRessoAggregate - for aggregating results from previously-run CRISPResso analyses

How can you use CRISPResso?

CRISPResso can be used to analyze genome editing outcomes using cleaving nucleases (e.g. Cas9 or Cpf1) or noncleaving nucleases (e.g. base editors). The following operations can be automatically performed:

Filtering of low-quality reads
Adapter trimming
Alignment of reads to one or multiple reference sequences (in the case of multiple alleles)
Quantification of HDR and NHEJ outcomes (if the HDR sequence is provided)
Quantification frameshift/inframe mutations and identification affected splice sites (if an exon sequence is provided)
Visualization of the indel distribution and position (for cleaving nucleases)
Visualization of distribution and position of substitutions (for base editors)
Visualization of alleles and their frequencies

CRISPResso processing

CRISPResso Schematic

Quality filtering

Input reads are first filtered based on the quality score (phred33) in order to remove potentially false positive indels. The filtering based on the phred33 quality score can be modulated by adjusting the optimal parameters (see additional notes below).

Adapter trimming

Next, adapters are trimmed from the reads. If no adapter are present, select 'No Trimming' under the 'Trimming adapter' heading in the optional parameters. If reads contain adapter sequences that need to be trimmed, select the adapters used for trimming under the ‘Trimming adapter’ heading in the optional parameters. Possible adapters include Nextera PE, TruSeq3 PE, TruSeq3 SE, TruSeq2 PE, and TruSeq2 SE. The adapters are trimmed from the reads using fastp.

Read merging

If paired-end reads are provided, reads are merged using fastp. This produces a single read for alignment to the amplicon sequence, and reduces sequencing errors that may be present at the end of sequencing reads.

Alignment

The preprocessed reads are then aligned to the reference sequence with a global sequence alignment algorithm that takes into account our biological knowledge of nuclease function. If multiple alleles are present at the editing site, each allele can be passed to CRISPResso and sequenced reads will be assigned to the reference sequence or origin.

Visualization and analysis

Finally, after analyzing the aligned reads, a set of informative graphs are generated, allowing for the quantification and visualization of the position and type of outcomes within the amplicon sequence.

How is CRISPResso2 different from CRISPResso?

CRISPResso2 introduces four key innovations for the analysis of genome editing data:

Comprehensive analysis of sequencing data from base editors. We have added additional analysis and visualization capabilities especially for experiments using base editors.
Allele specific quantification of heterozygous references. If the targeted editing region has more than one allele, reads arising from each allele can be deconvoluted.
A novel biologically-informed alignment algorithm. This algorithm incorporates knowledge about the mutations produced by gene editing tools to create more biologically-likely alignments.
Ultra-fast processing time.

Installation

CRISPResso can be installed using the conda package manager Bioconda, or it can be run using the Docker containerization system.

Bioconda

To install CRISPResso using Bioconda, download and install Anaconda Python, following the instructions at: https://docs.anaconda.com/free/anaconda/install/.

Open a terminal and type:

conda config --add channels defaults
conda config --add channels bioconda
conda config --add channels conda-forge

To install CRISPResso into the current conda environment, type:

conda install crispresso2

Alternatively, to create a new environment named crispresso2_env with CRISPResso, type:

conda create -n crispresso2_env -c bioconda crispresso2

Activate your conda environment:

conda activate crispresso2_env

Verify that CRISPResso is installed using the command:

CRISPResso -h

Bioconda for Apple Silicon

If you would like to install CRISPResso using bioconda on a Mac with Apple silicon (aren't sure?), then there is a slight change you need to make. First, ensure that you have Rosetta installed. Next, you must tell bioconda to install the Intel versions of the packages. If you would like to do this system wide, which we recommend, run the command:

conda config --add subdirs osx-64

Then you can proceed with the installation instructions above.

If you would like to use the Intel versions in a single environment, then run:

CONDA_SUBDIR=osx-64 conda create -n crispresso2_env -c bioconda crispresso2

If you choose to use the CONDA_SUBDIR=osx-64 method, note that if you install additional packages into the environment you will need to add the CONDA_SUBDIR=osx-64 to the beginning of each command. Alternatively, you could set this environment variable in your shell, but we recommend to use the conda config --add subdirs osx-64 method because it is less error prone.

Docker

CRISPResso can be used via the Docker containerization system. This system allows CRISPResso to run on your system without configuring and installing additional packages. To run CRISPResso, first download and install docker: https://docs.docker.com/engine/installation/.

Next, Docker must be configured to access your hard drive and to run with sufficient memory. These parameters can be found in the Docker settings menu. To allow Docker to access your hard drive, select 'Shared Drives' and make sure your drive name is selected. To adjust the memory allocation, select the 'Advanced' tab and allocate at least 4G of memory.

To run CRISPResso, make sure Docker is running, then open a command prompt (Mac) or Powershell (Windows). Change directories to the location where your data is, and run the following command:

docker run -v ${PWD}:/DATA -w /DATA -i pinellolab/crispresso2 CRISPResso -h

The first time you run this command, it will download the Docker image. The -v parameter mounts the current directory to be accessible by CRISPResso, and the -w parameter sets the CRISPResso working directory. As long as you are running the command from the directory containing your data, you should not change the Docker -v or -w parameters.

Additional parameters for CRISPResso as described below can be added to this command. For example,

docker run -v ${PWD}:/DATA -w /DATA -i pinellolab/crispresso2 CRISPResso -r1 sample.fastq.gz -a ATTAACCAAG

Troubleshooting

Please check that your input file(s) are in FASTQ format (compressed fastq.gz also accepted).

If you get an empty report, please double check that your amplicon sequence is correct and in the correct orientation. It can be helpful to inspect the first few lines of your FASTQ file - the start of the amplicon sequence should match the start of your sequences. If not, check to see if the files are trimmed (see point below).

It is important to determine whether your reads are trimmed or not. CRISPResso2 assumes that the reads ARE ALREADY TRIMMED! If reads are not already trimmed, select the adapters used for trimming under the ‘Trimming Adapter’ heading under the ‘Optional Parameters’. This is FUNDAMENTAL to CRISPResso analysis. Failure to trim adaptors may result in false positives. This will result in a report where you will observe an unrealistic 100% modified alleles and a sharp peak at the edges of the reference amplicon in figure 4.

The quality filter assumes that your reads uses the Phred33 scale, and it should be adjusted for each user’s specific application. A reasonable value for this parameter is 30.

If your amplicon sequence is longer than your sequenced read length, the R1 and R2 reads should overlap by at least 10bp. For example, if you sequence using 150bp reads, the maximum amplicon length should be 290 bp.

Especially in repetitive regions, multiple alignments may have the best score. If you want to investigate alternate best-scoring alignments, you can view all alignments using this tool: http://rna.informatik.uni-freiburg.de/Teaching/index.jsp?toolName=Gotoh. As input, sequences from the 'Alleles_frequency_table.txt' can be used. Specifically, for a given row, the value in the 'Aligned_Sequence' should be entered into the 'Sequence a' box after removing any dashes, and the value in the 'Reference_Sequence' should be entered into the 'Sequence b' box after removing any dashes. The alternate alignments can be selected in the 'Results' panel in the Output section.

Changelog

Unreleased

ADDED

Add an amino acid nucleotide quilt plot by @mbowcut2 in #552
Add scripts/reconstituteReads.py to generate FASTQ from CRISPResso2 output by @kclem in a800762 and cd79dcc
Add an UpSet plot to represent bystander edits for Base Editing analyses by @mbowcut2 in #554
Allow for messages to be served via CRISPResso reports by @Colelyman in #583
Add a plot that shows the distribution of homology scores for reads by @mbowcut2 in #600

FIXED

Fix the x_lim settings on plot 3b by @kclem in 56bd430
Fix parsing the CRISPResso2_info.json in CRISPRessoPooled by @kclem in #558
Forced cloned include_idxs to be np.arrays by @kclem in da4badb
Fix the link to the CRISPResso cup in reports (so that SSL works correctly) by @Colelyman in #571
Fix the quantification of deletions at the second position of the sequence by @Colelyman in #574
Fix an issue with unaligned reads not being reported correctly when writing BAM output by @trevormartinj7 in #578
Fix an issue where quantification window coordinates we not being correctly inferred by @Colelyman in #598
- This issue is present when there is a single quantifcation window coordinate provided and multiple amplicons. What happens is CRISPResso aligns the second amplicon to the first and then infers what the quantification window coordinates should be based on the alignment. A regression was introduced where the inference of the quantification window coordinates for the second amplicon was no longer correct. This change fixes the regression and brings the behavior back to match that of v2.2.9.
- If you don't set quantification window coordinates and don't use multiple amplicons, there is no need for this fix and therefore no change in behavior.
Fix a SyntaxWarning for an unescaped sequence in a matplotlib function by @Colelyman in #600
Fix a bug during --bam_output when there is an unaligned read, the remainder of the reads will not bu output by @Colelyman in #602

Keyboard shortcuts

CRISPResso Docs