How to use PASPA

Suggested Browsers

Advanced browsers, such as Chrome, Firefox, Safari, and Internet Explorer (10.0 or later) can be used for browsing PASPA.
Firefox and Chrome are recommended:
Mozilla Firefox 3.5 or greater (
Google Chrome 6.0 or greater (

1. Introduction

PASPA is a web server for poly(A) site prediction in plants and algae. Currently, PASPA can predict poly(A) sites for ten species, including Arabidopsis, rice, and Medicago truncatula, spikemoss Selaginella moellendorffii, moss Physcomitrella patens, red algae Cyanidioschyzon merolae, two green algae C. reinhardtii and Ostreococcus lucimarinus, and two diatoms Thalassiosira pseudonana and Phaeodactylum tricornutum.

2. Inputs for PASPA

2.1 Start a new task for poly(A) site prediction

1) Choose species

The supported species are listed in the drop-down box. However, if the species desired is not listed, you can choose the similar species in the list instead or the program can automatically decide the optimal parameter file to proceed with your poly(A) site prediction (when you choose “unknown”).

2) Input sequences

PASPA provides two ways to input sequences. You can either upload a file in FASTA format or paste sequences in the text box. You can also click “Try an example” button to load the demo data.

■ FASTA file (recommended)

■ Copy and paste

3) Upload file with real poly(A) sites (optional)

You can upload a file to specify real poly(A) sites in your input sequences. If this file is provided, PASPA will validate the predicted poly(A) sites by the real ones. Please click the “Example” link to view the format of this file.

4) Options for locating poly(A) sites (optional)

PASPA can determine the exact position of predicted poly(A) sites. Users can specific the minimum distance between two predicted poly(A) sites (default: 30 nt) and the total number of predicted poly(A) sites (default: 6) for locating poly(A) sites. Of note, different values of these two options may result in different predicted poly(A) sites, while users can view the score curve in the score viewer to explore potential regions of predicted poly(A) sites.

2.2 Search a task by task ID

Each time, when a user visits our website to conduct a poly(A) site prediction, an unique task ID will be assigned. In our server, a folder in the same name as this project name will be created to hold all relevant data for the poly(A) site prediction. The user can further obtain all prediction results of this task by searching the task ID in the “Get Results” page.

All tasks started from the same IP address are also listed in the “Get Results” page for the user to retrieve at any time.

3. Outputs for PASPA

The output is the poly(A) site(s) predicted along the original target sequences, based on the poly(A) scores. A user can download the files recording the predicted poly(A) sites and the GHMM scores of the target sequences to the local computer for further analysis. You can also download all figures with GHMM scores and predicted (and known) poly(A) sites plotted. The user can choose to visualize various poly(A) site prediction results using three viewers in PASPA, including list viewer, score viewer, and EST-PolyA viewer.

list viewerscore viewerEST-PolyA viewer

3.1 List viewer

1) Task Summary

The general output of PASPA is first given in the list viewer. The poly(A) site prediction results are provided in a tabulated list that holds the information about the given sequences, the known poly(A) sites, and the predicted poly(A) sites.
If a predicted site is in the vicinity of any known site (± 30 nt),then it is considered as valid.

2) Sequence statistics

Users may also choose to visualize detailed information of their input sequences, including single nucleotide base compositions, poly(A) signals of the input sequences, and statistics of the four bases.

3.2 Score viewer

By following the web link in the list of list viewer, a user can take advantage of the score viewer to visualize the PASS scores and signals of a selected sequence.

1) PASS score

In the figure of PASS score, all predicted and known poly(A) sites will be automatically anchored to the score series of the given sequence. Through our GHMM-based prediction model, each nucleotide of a give sequence will correspond to a score value denoting its possibility of being a poly(A) site. Thus, from the score curve, the regions where the potential poly(A) sites are most likely located can be clearly shown.

2) Pattern searching

Users can further choose to visualize poly(A) signals, the predicted poly(A) sites and known sites at a single-nucleotide resolution. A user can identify canonical poly(A) signals (e.g., AAUAAA and UGUAA) and their variants or search other patterns. You can also extract possible sub-sequences with alternative polyadenylation for other downstream analyses (e.g., BLAST and miRNA target prediction).

3.3 EST-PolyA viewer

By following the web link in the score viewer, the page will be directed to the EST-PolyA viewer to explore authenticated polyadenylation evidence.

The input sequences are mapped to the genome using an alignment program for mRNA and EST called GMAP. Then the aligned sequence is scanned for known poly(A) sites in the real poly(A) site databases. Currently, databases with highly confident poly(A) sites from large scale genomic studies for four species are available, including Arabidopsis, rice, M. truncatula, and C. reinhardtii. If a sequence can be aligned to the genome, the EST alignment, the predicted poly(A) sites, and the known sites (if available) can be visualized in an EST-genome plot.

An additional list is provided to tabulate the known poly(A) sites in a given range of the aligned sequence. Details of the known poly(A) sites are also recorded, such as strand, coordinate, number of the supporting reads.

The sequence alignment is also provided in a text box to display the alignment of each nucleotide of the input sequence. A user can extract possible sub-sequences of interest for further analysis.

3.4 Download output files

Users can download output files by clicking links in the bottom panel of the above three viewers.

1) Files from list viewer

In List viewer, Users can download the files recording the predicted poly(A) sites and the GHMM scores of the target sequences to the local computer for further analysis.

2) Files from score viewer

In score viewer, Users can download the files recording the predicted poly(A) sites and the GHMM scores of the selected sequence to the local computer.

3) Files from EST-PolyA viewer

In EST-PolyA viewer, Users can download the file recording alignments and the GMAP plot to the local computer.

© Copyright BMI Lab 2005-2013