NECTAR classifies genetic mutations into 14 classes based on the following three criteria:
I) consequence types
II) degree of damaging effect
III) is the mutation novel?
I. Mutation Consequence Types
This category is basically built on the basis of 'predicted variation consequences' of
Ensembl Variations with the help of
Ensembl Variation API.
There are
four levels:
- a: any from
- ESSENTIAL_SPLICE_SITE
- FRAMESHIFT_CODING
- STOP_GAINED
- STOP_LOST
- TRANSCRIPT_ABLATION
- b: any from
- c: any from
- SPLICE_SITE
- TRANSCRIPT_AMPLIFICATION
- d: none of above
II. Degree of Damaging Effects (level b only)
This category is for non-synonymous mutations (
b) where amino acids are changed or deleted. This is to answer 'how deleterious those missense and inframe deletion mutations are' by investigating structural and functional effects based on the three criteria as shown in the Venn diagram.
There are
four classes:
- 3: satisfying all the three coditions
- 2: satisfying any of the two conditions
- 1: satisfying any of one condition
- 0: satysfying none
Based on
I and
II above, there could be 7 possible combinations for a mutation class, which make the
mutation signal system.
III. Is the Mutation Novel?
Finally, NECTAR tells you whether the mutation has been reported previously based on dbSNP and HGMD.
It is encoded
1 for novel (not reported) and
0 for seen, which is added as the last digit to the mutation code.
For example, a mutaion of '
a11' stands for:
- a for category I
- 1 for category II
- 1 for category III (novel)