What are those "ENOG" numbers in eggNOG 4.5?

Question

[Dgg32]

Hello.

I noticed that eggNOG has updated to version 4.5. I downloaded their NOG.members.tsv from the ftp. And all the sudden, there are these "ENOG" numbers everywhere, examples:

ENOG41118X1
ENOG4111BGB
ENOG410YBUW

It seems that the COG classification loves to create new categories, from COG to KOG to NOG and now to ENOG.

I would like to ask, does anybody have an idea, what are these new ENOGs numbers? I search several of them in eggNOG, the returning results bear no difference with their COG counterparts. So any hints are appreciated.

Also, I have old COG annotations from http://www.string-db.org. There are still NOG numbers in them. If I search NOG numbers now in eggNOG 4.5, (for example: NOG18999), it will say no results. So how can I convert the old NOG number to the newer COG/ENOG system please? Thanks!

Answer 1

[jhc]

Original COG, KOG and arCOG IDs are still maintained in eggNOG 4.5 (example).

The nomenclature of NOGs (non supervised orthologous groups) changed from NOGxxxxxx to ENOG41xxxxx. The "E" identifies the group as eggNOG-produced. 41 identifies the internal eggnog release version. The next 5 characters are unique across eggnog versions and identify the actual orthologs group (OG). You could actually remove the ENOG41 tag and still search the OG in eggNOG.

Cross mapping of OG names between eggnog versions is not available at the moment. It was also not possible in previous versions. For instance NOG18999 in eggnog2 is not the same as NOG18999 in eggnog3. If you need to find NOG18999 in eggnog4.5, you will need to search for any representative protein contained in such group (NOG18999).

[Dgg32]

Just the NOG numbers, or even the COG numbers are not consistent across versions any more please?

[jhc]

just the NOG, COGs, arCOGs and KOGs are stable

Answer 2

[edward.messick]

extending this question, is there any database/flat file that has functional categories and/or associated gene/protein to each ENOG value? this exists for COGs but now that you have pointed out that COG's change between versions of eggNOG, I am no longer sure that these flat files are even usable... here is an example: ftp://ftp.ncbi.nih.gov/pub/wolf/COGs/COG0303/cogs.csv

[jhc]

http://eggnogdb.embl.de/#/app/downloads

[edward.messick]

awesome! thanks a lot - exactly what I need