Off-Target Site Finder identifies candidate binding sites for a TAL effector in a genome or promoterome of interest. Off-Target Site Finder is intended to identify binding sites for custom TAL effectors in the genome/promoterome.

TAL effector binding sites are scored using the scoring function developed by Moscou and Bogdanove in supplementary script S1 of the paper Moscou, M.J. and Bogdanove, A.J. (2009) A simple cipher governs DNA recognition by TAL effectors. Science. 326(5959):1501.

The scoring function is based on RVD-nucleotide association frequencies for known TAL-effector target pairs. Each RVD-nucleotide pair in the TAL effector/target alignment is assigned a probability score based on these association frequencies. Scores for all RVD-nucleotide pairs are summed to score the entire alignment.

The Off-Target Site Finder returns a list of the best (lowest) scoring candidate binding sites in a genome or promoterome for the TAL effector.

1. Select the genome/promoterome and enter the RVD sequence you want to search.
   To follow along with this tutorial, use the link at the top of the page to select a genome/promoterome sequence and load the sample RVD sequence:

   

   To enter a different RVD sequence, enter a sequence between 12 and 35 RVDs, separated by spaces. Use '*' to indicate a missing amino acid (such as N* or H*).

2. Select additional options.
   You may choose to scan the reverse complement of the genome/promoterome. By default only the forward sequence (the sequence exactly as it is downloaded in fasta format) is searched. To follow along with the tutorial, leave this box unchecked (default settings).
3. Submit your query.
   Optionally, you may enter your email address to receive an email with a link to your results when your job is complete.

   You must also choose the length of time your results will be available for download from our server.

   

   When you are finished, hit the Submit button.

4. Retrieve your results.
   After hitting Submit, you will be taken to a page detailing the progress of your job.

   

   You can hide the progress details by clicking on the arrow next to "Process log".

   Do not navigate away from this page; or bookmark the page so you can return and download your results later! Your job may take several minutes to finish.

   When the job finishes, your results will appear in a table. You will also receive a link to download your results as tab-delimited text. If you entered an email address, you will receive an email notification with a link to this page so you can retrieve your results.

5. Interpret your results.
   Off-Target Site Finder will return all sites in the genome/promoterome scoring below (better than) a cutoff score. The cutoff score is based on the Best Possible Score returned above the results table or at the top of the output file.

   Best possible score gives the score for the TAL effector on its "perfect" binding site (the site with all RVDs aligned with their most frequently associated nucleotide). If a site's Score is closer to the Best possible score, the TAL effector is more likely to bind to that site.

   

   TAL effectors in nature and their known targets typically have Scores less than 2-3 times the Best possible score for the TAL effector. Therefore, all sites scoring below 3*Best_possible_score are returned in the output. Not all sites listed are guaranteed to be bound by the TAL effector and some binding sites that score above the cutoff may be missed. However, the list of sites should contain the majority of binding sites in the genome interest, with a longer list indicating that a TAL effector is more likely to have off-target binding sites.

   You may wish to filter your list in order to address a question of interest. For example, when searching for additional off-target sites in a promoterome, you may be interested in sites that will result in gene activation. Therefore, you may want to consider only sites that are near a gene's transcriptional or translational start site.

   Note that all binding sites are directly preceded by a T at the 5' end. The T is not shown in the output.