Replies: 2 comments 2 replies
-
The RGroup decomposition will only give you one match here. It is designed
to keep the same rgroup fragments together as much as possible when doing a
decomposition, in this case since you had only one molecule, it will be an
arbitrary match.
Also note that the decomposition will look weird because you are splitting
a ring. I have some code that makes it prettier I can post in a bit.
…On Tue, Mar 26, 2024 at 7:48 AM Julianus Pfeuffer ***@***.***> wrote:
Hi!
I would like to enumerate all possible R-group decompositions for a given
core, not only one. Is this possible with the built-in functionality?
Core (rgd_test.sdf):
Mrv2311 03262410502D
12 10 0 0 1 0 999 V2000
-3.2067 0.3583 0.0000 H 0 0 0 0 0 0 0 0 0 0 0 0
-2.4943 -0.0577 0.0000 O 0 0 0 0 0 0 0 0 0 0 0 0
-1.7778 0.3512 0.0000 C 0 0 0 0 0 0 0 0 0 0 0 0
-1.7736 1.1762 0.0000 O 0 0 0 0 0 0 0 0 0 0 0 0
2.9221 -0.0243 0.0000 R# 0 0 0 0 0 0 0 0 0 0 0 0
1.4976 -0.8573 0.0000 O 0 0 0 0 0 0 0 0 0 0 0 0
2.2054 0.3841 0.0000 O 0 0 0 0 0 0 0 0 0 0 0 0
1.4933 -0.0323 0.0000 C 0 0 0 0 0 0 0 0 0 0 0 0
0.7764 0.3762 0.0000 N 0 0 0 0 0 3 0 0 0 0 0 0
0.4018 1.1113 0.0000 R# 0 0 0 0 0 0 0 0 0 0 0 0
0.1930 -0.2072 0.0000 R# 0 0 0 0 0 0 0 0 0 0 0 0
-1.0654 -0.0649 0.0000 R# 0 0 0 0 0 0 0 0 0 0 0 0
2 1 1 0 0 0 0
3 2 1 0 0 0 0
4 3 2 0 0 0 0
7 5 1 0 0 0 0
8 7 1 0 0 0 0
9 8 1 0 0 0 0
9 10 1 0 0 0 0
9 11 1 0 0 0 0
8 6 2 0 0 0 0
3 12 1 0 0 0 0
M RGP 4 5 5 10 2 11 3 12 1
M END
$$$$
ms = [x for x in Chem.SDMolSupplier('./rgd_test.sdf')]ms += [Chem.MolFromSmiles('[H]OC(=O)OC1=CC(=CN(C1)C(=O)OCCC)C(=O)O[H]')]res,_ = rdRGroupDecomposition.RGroupDecompose([ms[0]],ms[1:])for row in res:
print(f' {row}')
Draw.MolsToGridImage(res[0].values(),legends=res[0].keys())
Result: Only one decomposition
Expected result: My assumption is, that two are possible (for different
matches of R1).
I tried some RGD params but none worked so far.
I know it's probably not your typical decomp but it would make my life
much easier if this were possible.
Thanks
—
Reply to this email directly, view it on GitHub
<#7297>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACLOFIANGO66J6RLXQ3GN3LY2FHAPAVCNFSM6AAAAABFIYMKSGVHI2DSMVQWIX3LMV43ERDJONRXK43TNFXW4OZWGQZDENRVHA>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
1 reply
-
What do you mean by "all possible r-group decompositions"? Internally the code has the option to enumerate all symmetry matches and then scores the decompositions to pick one that best fits the entire set. Would you like this exposed or simply know how to do it? |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Hi!
I would like to enumerate all possible R-group decompositions for a given core, not only one. Is this possible with the built-in functionality?
Core (rgd_test.sdf):
Result: Only one decomposition
Expected result: My assumption is, that two are possible (for different matches of R1).
I tried some RGD params but none worked so far.
I know it's probably not your typical decomp but it would make my life much easier if this were possible.
Thanks
Beta Was this translation helpful? Give feedback.
All reactions