Pushed to juanlmateo/cctop_standalone
95ea199 Added CRISPRater score (doi: 10.1093/nar/gkx1268). Now the off-target site score is also computed even if the gene/exon files are not present. Some bugs fixed.
CCTop is a tool to determine suitable CRISPR/Cas9 target sites in a given query sequence(s) and predict its potential off-target sites. The online version of CCTop is available at http://crispr.cos.uni-heidelberg.de/ This is a command line version of CCTop that is designed mainly to allow search of large volume of sequences and higher flexibility. If you use this tool for your scientific work, please cite it as: Stemmer, M., Thumberger, T., del Sol Keyer, M., Wittbrodt, J. and Mateo, J.L. CCTop: an intuitive, flexible and reliable CRISPR/Cas9 target prediction tool. PLOS ONE (2015). doi:10.1371/journal.pone.0124633 REQUIREMENTS CCTop is implemented in Python and it requires a version 2.7 or above. In addition we relay on the short read aligner Bowtie 1 to identify the off-target sites. Bowtie can be downloaded from this site http://bowtie-bio.sourceforge.net/index.shtml in binary format for the main platforms. You need to create an indexed version of the genome sequence of your target species. This can be done with the tool bowtie-build included in the Bowtie installation. For that you simply need a fasta file containing the genome sequence. To get the index you can do something like: $ <path-to-bowtie-folder>/bowtie-build -r -f <your-fasta-file> <index-name> The previous line will create the index files in the current folder. To handle gene and exon annotations we use the python library bx-python (https://bitbucket.org/james_taylor/bx-python/). This library is only required if you want to associate off-target sites with the closest exon/gene, otherwise you don't need to install it. Notice, however, that in this case all candidate target sites will be given the same score, because in the current version the score of the candidate target sites considers only off-target sites with associated exon/gene. The exon and gene files contain basically the coordinates of those elements in bed format (http://genome.ucsc.edu/FAQ/FAQformat#format1), which are the first three columns of the file. There are two more columns with the ID and name of the corresponding gene and a sixth empty column to comply with the format accepted by the library. You can generate easily such kind of files for you target organism using Ensembl Biomart (http://www.ensembl.org/biomart). In case of difficulties with these files contact us and we can provide you the ones you need or help you to generate you own. INSTALLATION Simply download the two .py files (CCTop.py and bedInterval.py) to a folder of your choice and follow the instructions that you can find in the respective web sites to install Bowtie 1 and bx-python. USAGE You can run CCTop with the -h flag to get a detailed list of the available parameters. For instance: $ python <download-folder>/CCTop.py -h<Enter> At minimum it is necessary to specify the input (multi)fasta file (--input) and the index file (--index). In this case CCTop assumes that the Bowtie executable can be found in the current folder, there are not gene and exon file to use and the rest of parameters will take default values. Notice that the index file to specify refers to the name of the index you specified for bowtie-build together with the path, if necessary. A command for a typical run will look something like this: $ python <download-folder>/CCTop.py --input <query.fasta> --index <path/index-name> --bowtie <path-to-bowtie> --output <output-folder> <Enter> The result of the run will be three files for each sequence in the input query file. These files will have extension .fasta, .bed and .xls, containing, respectively, the sequence of the target sites, their coordinates and their detailed information as in the online version of CCTop. The name of the output file(s) will be taken from the name of the sequences in the input fasta file.