N-SCAN / Twinscan Gene Predictor

Before gene prediction, the sequence should be masked for interspersed repeats. These repeats are degenerate copies of transposable elements, and make up about a third of the human genome. The protein coding portion of a gene almost never contains interspersed repeats, therefore masking results in better gene prediction.

Low complexity and simple repeats are short, repetitive sequences such as TATATATATA or GAGATAGAGAGA. Genes sometimes do contain such repeats, so by default they are not masked. If you see whole exons consisting of such repeats in your results, you may want to re-run prediction with the simple repeats masked.

The default settings for masking are mask interspersed, don't mask low complexity and simple repeats. This is the way we run N-SCAN for whole genomes.

If you want to mask out some regions in your sequence, you can input them in lowercase and check the Mask Lowercase box. You can still select the other boxes depending on whether you want additional masking for repeats.

NSCAN uses organism-specific parameter files. If your organism is not in the list, we do not have a parameter file. In that case, if your organism is close to one that does appear in the list, select it. For example, if your input sequence comes from a fox, pick dog.

If nothing in the list comes close, pick the clade that contains your organism (e.g. 'vertebrate'), then select 'other...' in the Organism list. For instance, if your sequence is from elephant, pick 'other mammal'. If your sequence is Orang Utan, pick 'other primate'.