Chemical Structure Search Engines in Cyberspace Klaus Gubernator, Craig James, e. Molecules Inc. ACS 232 nd National Meeting Division of Chemical Information San Francisco, September 14, 2006
Chemistry on the Internet The web has revolutionized the way we retrieve information Chemistry is a late participant in this revolution
Search Google Images for “Aspirin” N
Structure of “triazolo thiadiazine” http: //scripts. iucr. org/cgi-bin/paper? cnor=a 12172&buy=yes Acta Cryst. (1975). B 31, 1427 -1429 The crystal structure of 7 -amino-2 H, 4 H-vic-triazolo[4, 5 -c]-1, 2, 6 thiadiazine 1, 1 -dioxide (ATT) C. Foces-Foces, F. H. Cano and S. García-Blanco Buy online You may purchase this article in PDF and/or HTML formats. For purchasers in the UK, and for purchasers elsewhere in the European Community who do not have a VAT number, VAT will be added to the price of the article. Format* PDF (US $40, plus US $7 for EC purchases)
Cheminformatics. org Datasets (which are, in contrast to other dataset lists, available in a structural format) This list will be expanded continuously. Please don't hesitate to make published datasets publicly available here. Currently available: 44 Datasets Note: The Briem/Lessel and Hert/Willett Dataset are only available as MDDR ID's due to license reasons. Please contact MDL for further information on the database. The datasets have nonethless been included here because they are standard datasets for similarity searching. – Andreas Bender Binary (active/inactive) datasets QSAR datasets QSPR datasets Toxicity datasets Metabolism datasets Permeability datasets Docking datasets Mechanistic datasets Mixed/Other datasets
Stahl dataset CS(=O)Nc 1 ccc(cc 1 OC 2 CCCCC 2)N(=O)=O CS(=O)Nc 1 cc 2 CCC(=O)c 2 cc 1 Oc 3 ccc(F)cc 3 F CS(=O)Nc 1 cc 2 CCC(=O)c 2 cc 1 Sc 3 ccc(F)cc 3 F CS(=O)Nc 1 ccc(cc 1 Sc 2 ccc(F)cc 2 F)C(=O)N CS(=O)Nc 1 ccc(cc 1 Sc 2 ccc(Cl)cc 2 Cl)S(=O)N COc 1 ccc(cc 1)c 2 sc(nc 2 c 3 ccc(cc 3)S(=O)C)c 4 ccccc 4 Cl COc 1 ccc(cc 1)c 2 sc(nc 2 c 3 ccc(SC)cc 3)c 4 ccccc 4 Cl CS(=O)c 1 ccc(cc 1)n 2 nc(cc 2 c 3 ccc(F)cc 3)C(F)(F)F CS(=O)c 1 ccc(cc 1)n 2 nc(cc 2 c 3 ccc(Br)cc 3)C(F)(F)F Cc 1 ccc(cc 1)c 2 cc(nn 2 c 3 ccc(cc 3)S(=O)N)C(F)(F)F CS(=O)c 1 ccc(cc 1)c 2 snnc 2 c 3 ccc(F)cc 3 CC(=O)c 1 nc(c(o 1)c 2 ccc(c(F)c 2)S(=O)N)c 3 ccccc 3 Cc 1 nc(C 2 CCCCC 2)c(o 1)c 3 ccc(c(F)c 3)S(=O)N CS(=O)c 1 ccc(cc 1)c 2[n. H]c(nc 2 C 3 CCCCC 3)C(F)(F)F CS(=O)c 1 ccc(cc 1)c 2[n. H]c(nc 2 c 3 ccc(F)cc 3)C(F)(F)F CS(=O)c 1 ccc(cc 1)C 2=C(C(=O)OC 32 CC 3)c 4 ccccc 4 CS(=O)c 1 ccc(cc 1)C 2=C(C(=O)OC 32 CCCC 3)c 4 ccccc 4 CS(=O)c 1 ccc(cc 1)c 2 cnn(Cc 3 ccccc 3)c(=O)c 2 c 4 ccccc 4 CS(=O)c 1 ccc(cc 1)c 2 nn(Cc 3 ccccc 3)c(c 2 c 4 ccc(F)cc 4)C(F)(F)F NS(=O)c 1 ccc(cc 1)c 2 c(CO)onc 2 c 3 ccccc 3 CS(=O)c 1 ccc(cc 1)c 2 cc(Cl)nn 2 c 3 ccc(F)cc 3 NS(=O)c 1 ccc(cc 1)c 2 cc(nn 2 c 3 ccc(F)cc 3)C(F)(F)F NS(=O)c 1 ccc(cc 1)n 2 nc(cc 2 c 3 nc 4 cccc(F)c 4 s 3)C(F)F
Yokoyama dataset Unnamed -MTS- 06200418093 D 0 0. 00000 0 13 13 0 0 0 0 1 V 2000 0. 0180 -0. 0030 0. 0000 S 0 0 0 1. 7880 0. 0070 0. 0000 C 0 0 0 2. 4880 0. 0110 1. 2120 C 0 0 0 3. 8880 0. 0200 1. 2120 C 0 0 0 4. 5880 0. 0240 0. 0000 C 0 0 0 6. 0030 0. 0330 0. 0000 N 0 0 0 6. 6610 1. 1880 0. 0000 C 0 0 0 6. 0400 2. 2500 0. 0000 O 0 0 0 8. 1410 1. 1970 0. 0000 C 0 0 0 8. 7570 0. 1440 0. 0000 O 0 0 0 8. 7890 2. 3360 0. 0000 O 0 0 0 3. 8880 0. 0200 -1. 2130 C 0 0 0 2. 4880 0. 0120 -1. 2120 C 0 0 0 1 2 1 0 0 2 3 2 0 0 3 4 1 0 0 4 5 2 0 0 … M END > <BIO> 48. 00 $$$$
Search Genbank for “aattccgg”
C
C
Why is so little chemistry on the web? Tradition? Strong providers of subscription services? Searching for chemical structures is significantly more difficult than text searching? Chemical identifiers are not standardized?
Open Access Chemical Search Engines Pub. Chem - NIH Chem. Bank – Harvard ZINC – UCSF Chem. DB – UC Irvine Chem. Exper - Lausanne Chem. Finder – Cambridge. Soft
www. emolecules. com New Chemistry Search Engine A large database of publicly available molecular structures Launched November 2005 50, 000 searches per month, rapidly growing
www. emolecules. com Free chemistry search site for publicly available chemical information
Advanced Search Powerful features: hit list management union, intersect, subtract, difference manual selection export lists in many formats persistent hitlists
T O
Content: 16 M entries, 5. 6 M structures Academic and government databases NIST Web. Book Drug. Bank Protein Ligands Chemical suppliers 150 electronic catalogs included Future goal All publicly available chemical information
Why is it so fast? Novel chemical search engine technology Method represents a major departure from previously known algorithms - Molecular keys (MDL) - Fingerprints (Daylight) - Feature Trees (Bio. Solv)
Search engine technology Analyze each molecule for distinguishing structural features Generate all features algorithmically Normalize features and use them for indexing Result: very fast searches
Who is e. Molecules? Klaus Gubernator Craig A. James Rashmi Mistry
Summary Free for depositors and users Very fast search engine High quality user interface Rich functionality Complementary with other engines
Contact Information Klaus Gubernator klaus@emolecules. com Skype: emolecules +1 -858 -764 -1941
Скачать презентацию Chemical Structure Search Engines in Cyberspace Klaus Gubernator
9397e0da98b32ae44c19e8eea8f00014.ppt