RetroTector online Help


RetroTector online is a reduced, web-based version of RetroTector© (ReTe). ReTe is an extensive Java package for searching of endogenous retroviruses in genomes. If the power of ReTe online, or this documentation, seems insufficient for you, the full ReTe, or its documentation, may be what you need. They may be requested from Prof. Blomberg.

ReTe was designed for search in large sequences. In principle this goes for ReTe online too, though for practcal reasons it limits sequence length to 10 Mb. Background DNA and integration repeats are used in the process. Therefore, though ReTe online may be used to analyse short sequences, consisting of a provirus only, this is not optimal. Short sequences will normally be "padded" (see below) to improve performance.

Procedure outline

After data have been presented the following steps will be carried out:

  1. If the sequence is shorter than 37 kb and 'Add padding' is checked, 15 kb will be added at either end of it. These are random bases, except for a 6-base simulated short direct repeat.
  2. If the sequence is longer than 115 kb, it will be cut into 'chunks', 115 kb long with 15 kb overlap. A provirus may therefore be found twice.
  3. Nonretroviral repetitive sequences such as ALUs etc may interfere with the process, especially by masquerading as LTRs. Their number is reduced by dedicated modules ('Brooms'). Since these are species-specific, the 'Species' menu allows optimal choice for a number of species.
  4. A number of 'LTR candidates' will be identified. In general, only a minority of them will be accepted as likely LTRs in the later procedure.
  5. Hits by a number of known retroviral motifs are collected and combined into 'Chains' ie suggested proviruses, partial or entire. If the score of a Chain is >350, it is quite likely to be genuine.
  6. If there are enough motif hits within a gene, a 'Putein' (or 'EnvTrace', through a special procedure for the env gene), a suggested reconstruction of the encoded protein, is attempted.
  7. If the Chain score is high, a search is made for rather suspect entities, 'XXons' and 'Xons'. If you are really interested, see the ReTe documentation.
  8. The results are packed into a .zip file.
The Submit data view

Sequence data may be in raw or FASTA format. Sequence numbers among the data will be stripped off. The data may be entered either (choose which with Toggle form) as a file, using the 'Choose file' button, or by pasting in 'Process data'.
You can also enter only the GenBank accession number of the sequence. Please note that the accession number (ACCESSION) sometimes is different from the LOCUS number. Therefore, always use the ACCESSION number if you want to refer to a sequence in short hand form in the submit data window.

By unchecking 'Add padding' (only if you really know what you are doing) you bypass point 1 above.

With the 'Species' menu, choose something as close as possible (see point 3 above).

After submitting the sequence for analysis with 'Send file' ('Send data'), you may click on 'Show queue' to switch to:

The Job queue view

By default, this shows both current and old jobs. You may customize the list using the 'Sort on' and 'Show menus' (followed by 'Refresh').

The job list is largely explained at the right side of the window. With the 'Download' button you may download the zipped results to your computer. It contains a number of files. If you need a deeper understanding of them, get the ReTe documentation (see above). For most purposes, however, it will be sufficient do display the results with (there is a download link at the upper right if you have not got it)


This is a double-clickable Java program, which requires a Java runtime (included in  MacOSX, downloadable from for Windows and a number of other platforms).

When it has started, find and enter the result file using the 'Directory ^' and 'Choose directory' buttons. If there are several 'chunks' (see point 2 above) those where chains were found will be shown in boldface. When you enter a chunk directory, the results, if any, will be displayed graphically.

RetroTectorviewer displays information in two frames. In the upper (text) frame, the contents of the Chain parameter are displayed, followed by the DNA sequence corresponding to the Chain.

The lower, graphic frame displays positions in DNA at the top, and the following features, separately for the three reading frames if applicable:

The first three features are always shown if applicable. The others may be turned off or on using the parameter panel at the right, clicking or unclicking the relevant boxes. For some reason, changes may not take effect until the cursor is moved into the graphics frame.

If 'Manual marking' is checked, a section of a reading frame may be selected with two clicks and viewed.

If the cursor is put over a Putein, XXon or Xon, it will be shown in red, and some information about it displayed at the bottom. If the cursor is anywhere in the graphic frame, the DNA position corresponding (approximately) to its horizontal position is shown at the bottom. These functions sometimes hang in mysterious fashion, but a mouse click anywhere in the window will help. Clicking will activate the Puteinview module with the component.

Clicking the 'Save as file' button saves the graphics to a PNG, EPS or JPEG file, 'Save as file, A4' does the same but folds the graphics to fit into A4 size.

Clicking the rectangle denoting an LTR in the graphic frame, details of the LTR will be shown in a separate window.

Only one Chain at a time is displayed. There is a menu(s) to select which.