Search Tip: Knowing whether to use U or T when searching RNA sequences on STN


Advanced Search »

Introduction

When you are designing a search query for sequence information on STN, it is important to consider how specific databases index RNA sequences.

Extensive sequence information is available in:

DGENE (GENESEQ^TM)
CAS REGISTRY^SM
USGENE (USPTO Genetic Sequence Database)
PCTGEN (World Patent Application Biosequences)

These databases derive their sequence content from multiple sources, including journals, patents, and GenBank. RNA sequences in these databases are indexed with U or T, depending on the source of the data.

For example, the National Center for Biotechnology Information (NCBI) has a policy of converting uracil (U) residues to thymidine (T) for RNA sequences in GenBank. Other sources of sequence data, however, index RNA sequences with U.

This article provides descriptions and examples of how you can effectively search for RNA sequences in STN databases.

DGENE (GENESEQ)

DGENE contains nucleotide and peptide sequences from basic patent documents of 41 patent-issuing authorities. RNA sequences in DGENE are reported with U residues. Therefore, you should include Us in your RNA search query sequences.

=> FIL DGENE 
=> RUN GETSEQ GGGAAUACCA/SQSN

L6          22 GGGAAUACCA/SQSN

=> D SQIDE

L6    ANSWER 1 OF 22  DGENE  COPYRIGHT 2008 THOMSON REUTERS on STN
AN    ASQ24900  RNA       DGENE
NA    4 A; 2 C; 3 G; 0 T; 1 U; 0 Other
SQL   10
SEQ
        1 gggaauacca 
          ==========
HITS AT:  1-10

CAS REGISTRY

RNA sequence records in REGISTRY can include U or T, depending on their original source. RNA sequences from GenBank contain T, but those from non-GenBank sources contain U.

But, REGISTRY automatically helps with this. A subsequence query containing U allows for ambiguous matches on either T or U.

In the following example, two records are retrieved using U in the Subsequence search. The first contains an RNA sequence indexed with U that is found in a PCT application.

=> FIL REGISTRY
=> S UGAAGCGGAGCUGGAA/SQSN AND SQL=16

      3 UGAAGCGGAGCUGGAA/SQSN
        101308 SQL=16
L1           3 UGAAGCGGAGCUGGAA/SQSN AND SQL=16
=> S L1 AND RNA/CNS; D SQIDE 1-2
L1   ANSWER 1 OF 2  REGISTRY  COPYRIGHT 2008 ACS on STN 
RN   866168-32-3  REGISTRY
CN   RNA, (U-G-A-A-G-C-G-G-A-G-C-U-G-G-A-A) (9CI) (CA INDEX NAME)
OTHER NAMES:
CN   4: PN: WO2005092393 FIGURE: 1 claimed RNA
FS   NUCLEIC ACID SEQUENCE
SQL  16
PATENT ANNOTATIONS (PNTE):
Sequence |Patent        
Source   |Reference     
=========+==============
Not Given|WO2005092393  
         |claimed FIGURE
         |1             
SEQ      1 ugaagcggag cuggaa

The second record contains an RNA sequence indexed with T instead of U. The original source of this sequence is GenBank.

L1   ANSWER 2 OF 2  REGISTRY  COPYRIGHT 2008 ACS on STN 
RN   422620-34-6  REGISTRY
CN   RNA (human microRNA miR-115 gene,) (9CI)  (CA INDEX NAME)
OTHER NAMES:
CN   GenBank AF480513
FS   NUCLEIC ACID SEQUENCE
SQL  16
NA   5 a   2 c   7 g   2 t
SEQ      1 tgaagcggag ctggaa
           ========== ======

USGENE (USPTO Genetic Sequence Database)

USGENE includes sequence records from the USPTO and GenBank. RNA sequences from the USPTO are indexed with U. RNA sequences from GenBank are indexed with T.

Unlike REGISTRY, searches in USGENE for sequences containing U do not automatically result in ambiguous matches on T or U. Searching with U only retrieves sequences containing U. Searching with T only retrieves sequences containing T.

For comprehensive results, it is therefore necessary to conduct two searches: one with T and a second with U.

The following example shows a search in USGENE for an RNA sequence containing U. The retrieved record is from the USPTO and is indexed with U.

=> FIL USGENE
=> RUN GETSEQ GGGAAUACCA/SQSN
L1          20 GGGAAUACCA/SQSN
=> S L1 AND SQL=10; D BIB MTY SSO SEQ
L2    ANSWER 1 OF 1 USGENE COPYRIGHT 2008 SEQUENCEBASE CORP on STN 
AN    20080167258.17  RNA       USGENE
TI    Trans-excision-splicing ribozyme and methods of use
      (Published Application)
IN    Testa Stephen M. (Lexington, KY); Bell Michael A. 
      (Lexington, KY)
PA    UNIVERSITY OF KENTUCKY RESEARCH FOUNDATION (Lexington KY)
PI    US 20080167258      A1   20080710
AI    US 2007-723456           20070320
DT    Patent
MTY   RNA
SSO   NUCLEIC; USPTO; APPLICATION
SEQ
        1 gggaauacca
          ==========

The next example shows a search for the same RNA sequence with T instead of U. The retrieved record (indexed with T) is from GenBank (NCBI) with an earlier application date than the USPTO record shown above.

=> RUN GETSEQ GGGAATACCA/SQSN
L1           6437 GGGAATACCA/SQSN
=> S L1 AND SQL=10; D BIB MTY SSO SEQ
L2    ANSWER 1 OF 1 USGENE COPYRIGHT 2008 SEQUENCEBASE CORP on STN 
AN    7211661.17  RNA       USGENE
TI    Trans-excision-splicing ribozyme and methods of use (Patent)
IN    Testa Stephen M. (Lexington, KY); Bell Michael A. (Lexington, KY)
PA    University of Kentucky Research Foundation (Lexington KY)
PI    US 7211661          B1   20070501
AI    US 2003-730261           20031209
DT    Patent
MTY   RNA
SSO   NUCLEIC; NCBI; GRANTED
SEQ
        1 gggaatacca                                            
          ==========

PCTGEN (World Patent Application Biosequences)

PCTGEN covers nucleotide and amino acid sequence information as submitted by the patent applicant to the World Intellectual Property Organization (WIPO). Therefore, RNA sequences in PCTGEN records contain either U or T. Searching with U only retrieves sequences indexed with U. Searching with T only retrieves sequences indexed with T. For comprehensive results, it is necessary to conduct two searches, with U and with T, as you would in USGENE.

This example shows an RNA search in which the query sequence contains U residues.

=> FIL PCTGEN => RUN GETSEQ GGGAAUACCA/SQSN

L1 31 GGGAAUACCA/SQSN

=> D L1 ALL

AN 2005116250.507809 RNA PCTGEN

TI VIRAL AND VIRAL ASSOCIATED MIRNAS AND USES THEREOF

PA ROSETTA GENOMICS LTD

PI WO 2005116250 20070712

ED 20070718

DT Patent

ORGN Human

SQL 59

SEQ

1 gacauaaaua agcauauuca gcaaaaaggu cugaaaagca agggaauacc

=========

51 auuauuuuc

HITS AT: 42-51

The following example shows a search for the same RNA sequence used in the example above, but U has been replaced with T. The records in the answer set are indexed with T.

=> RUN GETSEQ GGGAATACCA/SQSN
L2        1206 GGGAATACCA/SQSN

When the retrieved records are limited to those containing RNA sequences (RNA in the Molecule Type field (/MTY)), three records are found. These records would have been missed if the second search (with T) had not been conducted.

=> S L2 AND RNA/MTY

1589512 RNA/MTY

L3 3 L2 AND RNA/MTY

=> D ALL

AN 2004044123.543832 RNA PCTGEN

TI BIOINFORMATICALLY DETECTABLE GROUP OF NOVEL REGULATORY
GENES AND USES THEREOF

PA ROSETTA GENOMICS LTD

PI WO 2004044123 20040527

ED 20040529

DT Patent

ORGN Homo Sapiens

SQL 32

SEQ

1 tttaagaaaa tgtgggaata ccaaatgtcc tt

======= ===

HITS AT: 14-23

Conclusion

As more patents for DNA-related inventions are issued, comprehensive searches for sequence patentability become increasingly important. Freedom-to-operate, prior art, validity, and infringement patent sequence searches can be conducted in STN databases that contain extensive sequence and patent information. Keeping in mind the differences between these databases can help you improve your RNA sequence search strategies and retrieve more comprehensive results.

Overview of RNA sequence indexing.

	Thymidine (T)	Uracil (U)	Simultaneous Search
DGENE		X	n/a
REGISTRY	X	X	yes
USGENE	X	X	no
PCTGEN	X	X	no

Additional Resources

For additional information about the databases mentioned above, refer to the STN Database Summary Sheets.

For more information about sequence searching on STN, visit the Searching for Sequences documentation within STN Support & Training.

Updated: 10/9/2008 10:13:07 AM