[SO-devel] Changes for sequence_variation terms

[Karen E. <ei...@fr...>]

Hi

At the recent immunology workshop we decided to add new terms =20
(allele, genotype etc. ) to sequence_variant, which already had kinds =20=

of variation as its children (insertion, inversion etc). This has =20
caused some problems: allele and insertion should not be siblings.

One term is being used to describe  two phenomena; differences that =20
relate to populations, and differences between two sequences.

We need to modify the ontology.

The way I think this should be is that sequence_variant should be the =20=

term that is the parent of allele etc. It needs a new definition.

We need a new term sequence_alteration to capture the actual changes =20
in the sequence between a variant and the reference sequence. =20
(alteration is the neutral term suggested by HGVS.)

This would be the new structure:

sequence_variant

             allele

             genotype

             haplotype

             polymorphic_sequence_variant



sequence_alteration

             deletion =96 change definition

             deletion_junction - remove

             insertion

             insertion_site - remove

             inversion

             substitution

             translocation

             translocation_site - remove



These are my first stabs at defining these terms:

A sequence_variant is a non exact copy of a sequence_feature or =20
genome exhibiting one or more sequence_alteration.

A sequence_alteration is a sequence_feature whose extent is the =20
deviation from another sequence.

Comments, better definitions etc would be appreciated,

Karen=