Yes exactly. The 2-methylhexane match is what I tried to mean with the "linear, branched motif". Your picture is much clearer however. Just goes to show that a (well-made) picture is worth a thousand words!

When I tried to reproduce the results on my (outdated) SMSD copy, I noticed that that actually provides the best match too.
Here's the versions I tried (all settings left at default, same options selected in all versions):
SMSDgui20100718.jar --> matches 2-methylhexane (best)
SMSDgui-1.0.2.jar --> matches 2-methylhexane (best)
SMSDgui20100924.jar (current) --> matches hexane (sub-optimal)

So somewhere between July 18th and September 24th, the algorithm changed. This is also reflected in window titles, the first two show the results in "SMSD Maxmimum[sic] Common Substructure", the current version has "Isomorphism Maxmimum[sic] Common Substructure".

By the way Gilleain, what software is that you use to make those pictures?

Cheers! Jules

On 3 January 2011 18:03, gilleain torrance <gilleain.torrance@gmail.com> wrote:
Hi,

I think that the match shown in my image is a valid subgraph for both
structures. However, it doesn't look like the maximum common subgraphs
(see image). Imposed on top of the original images - where the blue
circles indicated the actual matched atoms - are red lines to show
connected subgraphs. The top shows the hexane subgraph that agrees
with the atom matches, and the bottom shows a larger 2-methylhexane
subgraph.

gilleain

------------------------------------------------------------------------------
Learn how Oracle Real Application Clusters (RAC) One Node allows customers
to consolidate database storage, standardize their database environment, and,
should the need arise, upgrade to a full multi-node Oracle RAC database
without downtime or disruption
http://p.sf.net/sfu/oracle-sfdevnl
_______________________________________________
Cdk-devel mailing list
Cdk-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/cdk-devel