Using sequence case to encode intron/exon information

Question

[anderspitman]

I've heard that sometimes sequences are stored in files partially lowercase and partially uppercase, with the intent that lower- or upper-case regions indicate introns or exons.

Is this a common practice? Are there file formats that explicitly support this type of encoding? Google hasn't yielded anything, but I might just be searching for the wrong things.

Answer 1

[h.mon]

Yes it is common practice, but it is not a standard.

Another common usage for lower case is to indicate softmasks, that is, lower case sequences should not be processed/analyzed.

[anderspitman]

Can you link to any examples where the convention is explained?

[h.mon]

Lower-case as exon-intron: this paper, which refers to this (or this) database; or see this server to generate pretty graphics of gene structure.

Lower case for soft-masking: see USEARCH manual or Blast lcase_masking parameter.