TargetScan input

[NicoBxl]

Hi,

I want to use TargetScan locally on my computer. So I downloaded TargetScan ( http://www.targetscan.org/cgi-bin/ta...cgi?db=vert_50) .

The input is composed like this :

The script takes two input files
1) A tab-delimited file that lists the miRNA seed sequences.
2) A tab-delimited multiple sequence alignment of the 3' UTRs of genes from the desired species.

The first point is ok. The second I don't know how produce such an alignment (what is the query ? ) . I've the 3' UTR of the organism I want to analyze ( in this research a virus ).

Anyone can help me ?

Thanks a lot,

Nicolas

[limits999]

Hi,

Have you solved this problem now?
I also want to know your methods about it at last.
Could you share your treatment?

Thanks,
Liao

[ejspina]

TargetScan Alignment File Creation From FASTA

After dabbling with several whole genome alignment programs looking for an easy way to output corresponding regions between two species, I came up with the following solution to create a multiple alignment file for input to TargetScan. It requires some simple scripting and command line use.

1) First, I downloaded all 3'utr sequences for both species as multi-fasta files from Ensemble Biomart. I also downloaded a file of matched sequence ID's (by homology) again using Ensemble Biomart. The Biomart attribute options didn't allow me to directly get orthologous transcripts, so I took orthologous proteins and changed the ID names (Ensemble transcript & protein ID's have the same numbers, they just differ in the T or P suffix). I don't know why, but there were duplicate transcript pairs in my file so I used MsExcel to remove them. I trimmed the file so each line contained only matched ID's for transcripts.
2) Next, I indexed the multi-fasta files into a coherent database structure using BioPythons SeqIO.index_db function.
3) After that I looped through lines of the ID list file to search the indexed database object, all entries found while looping through a line were written to a unique file (to make things easy I dumped the outputs into an empty directory). Rephrased, each iteration of my loop created a new multi-fasta file of ortholog sequences chosen from the larger multi-fasta files according to the list of matched IDs.
4) Then, I fed the ortholog pair multi-fasta files into an aligner (e.g Clustal) using a BASH for loop in Cygwin and output the results of the loop to a single file.
5) Lastly, I used a script to convert the ortholog transcript IDs to the ID for the ortholog from the main species I was interested in. This wouldn't be a problem if your sequences have proper gene names, but TargetScan requires orthologs in the alignment file to have identical ID's. I also printed the NCBI taxonomy ID to each line at this point. Make sure the output is tab separated and has UNIX style end of line characters (easy to do in Notepad++ if your script doesn't do it automatically).

[soltaniismael]

Error input

Hi,

I downloaded TargetScan script (Perl Script) from website, When runing the input files (A tab-delimited file that lists the miRNA seed sequences ) am I getting a syntax error (《Bareword found where operator expected at -e line 1, near “9A”》) and (Semicolon seems to be missing at...)
Anyone can help me how to repair this ? (exemple of input file if its possible )

thanks a lot