AKos
Consulting & Solutions GmbH
a chemoinformatics company

Home

Company Info

Impressum

Contact

Reoptimization of MDL Keys for Use in Drug Discovery

Joseph L. Durant,* Burton A. Leland, Douglas R. Henry, and James G. Nourse

MDL Information Systems, 14600 Catalina Street, San Leandro, California 94577

Received December 17, 2001

Abstract:

For a number of years MDL products have exposed both 166 bit and 960 bit keysets based on 2D descriptors. These keysets were originally constructed and optimized for substructure searching. We report on improvements in the performance of MDL keysets which are reoptimized for use in molecular similarity. Classification performance for a test data set of 957 compounds was increased from 0.65 for the 166 bit keyset and 0.67 for the 960 bit keyset to 0.71 for a surprisal S/N pruned keyset containing 208 bits and 0.71 for a genetic algorithm optimized keyset containing 548 bits. We present an overview of the underlying technology supporting the definition of descriptors and the encoding of these descriptors into keysets. This technology allows definition of descriptors as combinations of atom properties, bond properties, and atomic neighborhoods at various topological separations as well as supporting a number of custom descriptors. These descriptors can then be used to set one or more bits in a keyset. We constructed various keysets and optimized their performance in clustering bioactive substances. Performance was measured using methodology developed by Briem and Lessel. “Directed pruning” was carried out by eliminating bits from the keysets on the basis of random selection, values of the surprisal of the bit, or values of the surprisal S/N ratio of the bit. The random pruning experiment highlighted the insensitivity of keyset performance for keyset lengths of more than 1000 bits. Contrary to initial expectations, pruning on the basis of the surprisal values of the various bits resulted in keysets which underperformed those resulting from random pruning. In contrast, pruning on the basis of the surprisal S/N ratio was found to yield keysets which performed better than those resulting from random pruning. We also explored the use of genetic algorithms in the selection of optimal keysets. Once more the performance was only a weak function of keyset size, and the optimizations failed to identify a single globally optimal keyset. Instead multiple, equally optimal keysets could be produced which had relatively low overlap of the descriptors they encoded.

You can download the complete paper here.

In order to generate the keys you can modify your database using REXEC.

open DATABASE dbname/dbname TNAME dbname;

alter database drop TEXT2DKEYS;

alter database add TEXT2DKEYS NOSUBSET/SUBSET;

close database;

List of Remote Molecule Database 166 Key Set of Searchable Keys

The List of Searchable Keys

If you search a molecule database, there are 166 numbered keys that you can use. If you do not recognize the words for a specific key, the words might be in a special type of language called query language. The list of searchable keys is as follows:

Key	Description
1	ISOTOPE
2	103 < ATOMIC NO. < 256
3	GROUP IVA,VA,VIA PERIODS 4-6 (GE..)
4	ACTINIDE
5	GROUP IIIB,IVB (SC..)
6	LANTHANIDE
7	GROUP VB,VIB,VIIB (V..)
8	QAAA@1
9	GROUP VIII (FE..)
10	GROUP IIA (ALKALINE EARTH)
11	4M RING
12	GROUP IB,IIB (CU..)
13	ON(C)C
14	S-S
15	OC(O)O
16	QAA@1
17	CTC
18	GROUP IIIA (B..)
19	7M RING
20	Si
21	C=C(Q)Q
22	3M RING
23	NC(O)O
24	N-O
25	NC(N)N
26	C$=C($A)$A
27	I
28	QCH2Q
29	P
30	CQ(C)(C)A
31	QX
32	CSN
33	NS
34	CH2=A
35	GROUP IA (ALKALI METAL)
36	S HETEROCYCLE
37	NC(O)N
38	NC(C)N
39	OS(O)O
40	S-O
41	CTN
42	F
43	QHAQH
44	OTHER
45	C=CN
46	BR
47	SAN
48	OQ(O)O
49	CHARGE
50	C=C(C)C
51	CSO
52	NN
53	QHAAAQH
54	QHAAQH
55	OSO
56	ON(O)C
57	O HETEROCYCLE
58	QSQ
59	Snot%A%A
60	S=O
61	AS(A)A
62	A$A!A$A
63	N=O
64	A$A!S
65	C%N
66	CC(C)(C)A
67	QS
68	QHQH (&..)
69	QQH
70	QNQ
71	NO
72	OAAO
73	S=A
74	CH3ACH3
75	A!N$A
76	C=C(A)A
77	NAN
78	C=N
79	NAAN
80	NAAAN
81	SA(A)A
82	ACH2QH
83	QAAAA@1
84	NH2
85	CN(C)C
86	CH2QCH2
87	X!A$A
88	S
89	OAAAO
90	QHAACH2A
91	QHAAACH2A
92	OC(N)C
93	QCH3
94	QN
95	NAAO
96	5M RING
97	NAAAO
98	QAAAAA@1
99	C=C
100	ACH2N
101	8M RING OR LARGER
102	QO
103	CL
104	QHACH2A
105	A$A($A)$A
106	QA(Q)Q
107	XA(A)A
108	CH3AAACH2A
109	ACH2O
110	NCO
111	NACH2A
112	AA(A)(A)A
113	Onot%A%A
114	CH3CH2A
115	CH3ACH2A
116	CH3AACH2A
117	NAO
118	ACH2CH2A>1
119	N=A
120	HETEROCYCLIC ATOM>1 (&..)
121	N HETEROCYCLE
122	AN(A)A
123	OCO
124	QQ
125	AROMATIC RING>1
126	A!O!A
127	A$A!O>1 (&..)
128	ACH2AAACH2A
129	ACH2AACH2A
130	QQ>1 (&..)
131	QH>1
132	OACH2A
133	A$A!N
134	X (HALOGEN)
135	Nnot%A%A
136	O=A.1
137	HETEROCYCLE
138	QCH2A>1 (&..)
139	OH
140	O>3 (&..)
141	CH3>2 (&..)
142	N>1
143	A$A!O
144	Anot%A%Anot%A
145	6M RING>1
146	O>2
147	ACH2CH2A
148	AQ(A)A
149	CH3>1
150	A!A$A!A
151	NH
152	OC(C)C
153	QCH2A
154	C=O
155	A!CH2!A
156	NA(A)A
157	C-O
158	C-N
159	O>1
160	CH3
161	N
162	AROMATIC
163	6M RING
164	O
165	RING
166	FRAGMENTS