Lee H. Pratt

This query and viewing tool also requires JavaWebStart. You will find a link to Sun to obtain JavaWebStart when you first request this viewer. Details about obtaining JavaWebStart are on the general Instructions page.

Once JavaWebStart has been installed, you can then enter the viewer by launching the application.

Upon launching the program a single window is opened. It is necessary to select a cluster run.  "Sorghum ALL" at the top displays data derived from clustering all available sorghum sequences using TGICL from TIGR.  "NSF1" at the bottom displays a Milestone Uniscript Set described elsewhere on this web site.

After making this choice, a Gene Discovery window opens.  It provides you with four choices.

Cluster Overview presents visual information about the individual clusters.  Cluster Group 1 contains information about all clusters.  In the case of the Milestone Uniscript Set, you may also choose to view information about only single libraries or predefined groups of libraries from this list box.  Click the submit button to retrieve the data from our Oracle database.  Wait for the Progress completed message to appear and click OK.
The Expression Profile window allows you to scroll through the clusters visually.  If necessary, open the bottom of the window further in order to view the contig (cluster) numbers.  Each bar in this view represents a cluster.  You are starting with the singletons, so the bars are not very impressive.  You can change the vertical scale by changing the number in the Scale Y Axis box clicking on "Set Y Axis."  Scroll through the contigs with the scroll bar.  Reset the Y Axis by clicking on the box with this name.  Print the page at any time by clicking on the Print button. 
Click on a bar to bring up a "Contig Composition Pie Chart" window.  From this window, select "Group and Library Definition" for a bit more information about library and subgroup associations.  Select Expression by SubGroup to view data organized by subgroups, such as drought.  Select Contig Alignment to view alignment either at high resolution, showing actual base calls, or by clicking on the Zoom button at low resolution to view contig quality more readily.  Selecting Sequences in Contig and Consensus displays the names of individual ESTs in the contig as well as the consensus sequence.  Annotation displays information about the best hit to PIR.  Note that each column can be sorted by clicking on the column header.  Selecting View PIR Database reveals the alignment of the query sequence with that from PIR.  A user can sort and select sequences in any fashion desired and then click on "Sub Table" to bring up a new window containing only the selected sequences.  This process can be repeated as many times as one would like until the final selection is attained.
The Discovery Rate (number) window displayes the number of unique transcripts (uniscripts) as a function of the number of 3' ESTs clustered (red curve).  The slider bar can be used to identify the number of uniscripts at any desired number of 3' ESTs based on the best fit curve (green curve).  Alternatively, a number can be typed into the box.
The Discovery Rate (%) window displays the percentage of unique 3' ESTs (uniscripts) as a function of the number of 3' ESTs clustered.
The CTG Length window displays contig length as a function of contig number.  Because all ESTs in a contig are 3', this number is not expected to be too much longer than the longest single EST in a cluster.

Search Contig allows one to select a cluster group as for the Cluster Overview above and then identify contigs of interest.  A library or group of libraries (use the CTRL key on a PC) is selected.  One can then specify any combination of Contig Length, Ratio, Sequences/Contig/Library, or Sequences in Contig.  Alternatively, one can search directly for a Contig Number or the contig in which a particular sequence resides (Seq Name). By clicking on the Search button, one retrieves all data fitting the search criteria.  These data can be sorted in either ascending or descending fashion in any column by clicking on the column header.  For multiple sorts, sort first by one column, identify the contigs desired, then select Sub Table.  From the new table, repeat the process.  Continue until the desired outcome is attained.
Make Fasta - select one or more contigs, then clilck on this button to bring up a new window.  After making the choice offered, click on submit and continue in order to save a fasta file on your local computer.
Pie Chart, Show Sequences, Annotate the Contig, and Contig Alignment open the same windows already described above.

Search Subgroup is nearly identical to Search Contig, except that appropriate libraries have already been divided into subgroups. 

Search Annotation, after selecting a cluster group as previously described, allows one to search Protein Name, Seq Name and Species fields either singly or in combination.  Protein Name and Species refer to the BLAST return from PIR, while Seq Name refers to our own sequence designation.  Note that any field can be partial (e.g., EM in the Seq Name field will return all sequences from library EM1).  Show Sequences, Pie Chart, Annotate the Contig, Contig Alignment, and Sub Table are the same as already described above.