Polya_svm: prediction of mRNA polyadenylation sites by Support Vector Machine
INTRODUCTION:
This program takes a file containing DNA/RNA sequences in the FASTA format as input, and makes predictions for putative mRNA polyadenylation sites [or poly(A) sites] or generates results indicating the occurrences of different cis-elements. The program is implemented in PERL and runs under UNIX/LINUX systems. It requires the LIBSVM library (also included in the package). The current version of this program is 1.1.
DOWNLOAD & INSTALLATION:
1. Download the current version of Polya_svm (polya_svm_1.1.tar.gz) to a directory, e.g. /home/polya_svm/.
Windows Version: polya_svm.1.1.windows.zip
2. Extract the tar file to the directory by typing tar -zxvf polya_svm_1.1.tar.gz
3. A directory named "polya_svm" will appear. Switch to it by typing cd polya_svm
4. Type polya_svm.pl to run the program.
USAGE INSTRUCTIONS:
Usage: polya_svm.pl -i input [-o output] [-e cutoff] [-m mode] [-l location]
Options:
-i input input file containing sequence(s) in the FASTA format.
-o output output file with output format based on the running mode. Default is “junk.out”.
-e cutoff e-value cutoff for prediction. The lower the e-value, the more stringent for the prediction. The default cutoff is 0.5.
-m mode running mode:
if 'e', prediction mode, which returns prediction results. This is the default mode.
if 's', matching element mode, which returns matching results for 15 cis-elements.
-l location if a position number is given, the program will make only one prediction at the specified position.
OUTPUT FORMAT:
A. When the running mode is ‘e’, the default mode, the output file will include the following information:
1. Three marked lines indicating:
1) Number of sequences in the file
2) Number of sequences predicted to be positive.
3) The e-value used for prediction.
2. For each sequence, the predicted result will include:
1) sequence id
2) result line: +/- From To Max-Location E-value
+/- indicates positive or negative prediction for the sequence.
From and To indicates the high probability window.
Max_Location indicates the middle point of the peak region.
E-value is the e-value for the peak region.
NOTE:
1. The program does not make predictions for sequences shorter than 120 nt. Negative result is given in this case. it is too
short for prediction. No result is given.
2. Multiple positive regions are reported on different lines.
B. When the running mode is 's', the format file contains matching results for all cis-elements. The following symbols are used for similarity to the consensus:
+, highly similar, over the 75th percentile of all possible positive scores.
|, very similar, between the 50th-75th percentiles of all possible positive scores.
:, similar, between the 25th-50th percentiles of all possible positive scores.
., somewhat similar, 0-25th percentiles of all possible positive scores.
-, negative score.
C. When '-l location' is specified, the –100 to +100 nt region surrounding the position will be used to make a single prediction.
EXAMPLES:
1. Make poly(A) site prediction for the input sequences:
>polya_svm.pl -i test.fa -o test.out -m e
2. Find occurrences of 15 cis elements in the input sequences:
>polya_svm.pl -i test.fa -o test.out -m s
3. Make a poly(A) site prediction at the position 100:
>polya_svm.pl -i test.fa -o test.out -m e -l 100
REFERENCES:
Cheng, Y., Miura, R.M., and Tian, B. Prediction of mRNA polyadenylation sites by Support Vector Machine. Bioinformatics In press.
Hu, J., Lutz, C.S., Wilusz, J., and Tian, B. (2005). Bioinformatic identification of candidate cis-regulatory elements involved in human mRNA polyadenylation. RNA 11: 1485-1493.
Chang, C.-C. and Lin, C.-J. (2005) LIBSVM: a Library for Support Vector Machines
(www.csie.ntu.edu.tw/~cjlin/libsvm).
CONTACT:
Please contact Yiming Cheng (yc34@njit.edu) or Bin Tian (btian@umdnj.edu) for comments/suggestions.
Previous Versions:
polya_svm_1.0.tar