Using lexical analysis and parsing in genome research

• US 9,104,657 B2
• Filed: 05/02/2013
• Issued: 08/11/2015
• Est. Priority Date: 07/03/2012
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bowman, Stephen D., Reddy, Dandala V., Werts, David B.
• Primary Examiner(s): Saeed, Usmaan
• Assistant Examiner(s): Weinrich, Brian E.
• Application Number:
  US13/875,713
• Publication Number:
  US 20140012866A1
• Time in Patent Office: 831 Days
• Field of Search: 707/755, 702/20, 702/19
• US Class Current: 1/1
• CPC Class Codes: G06F 16/33 Querying
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G16B 20/00 ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
• CPC Class Codes: G16B 20/20 Allele or variant detection, e.g. single nucleotide polymorphism [SNP] detection
• CPC Class Codes: G16B 30/00 ICT specially adapted for sequence analysis involving nucleotides or amino acids
• CPC Class Codes: G16B 40/00 ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding

Abstract

Provided are techniques for locating one or more genome patterns. One or more lexical annotators that each identifies a sequence of nucleotides are created. One or more parsing rule annotators are created using at least one of (1) one or more of the lexical annotators, (2) one or more dictionary entries, and (3) one or more previously-defined parsing rule annotators. The one or more parsing rule annotators are used to discover the one or more genome patterns comprising a combination of the lexical annotators and the parsing rule annotators.

Claims

1. A method for locating a genome pattern, comprising:

creating, using a processor of a computer, one or more lexical annotators that each identify a sequence of nucleotides of nucleotide bases selected from A, C, G, and T;

providing (1) the one or more lexical annotators, (2) one or more dictionary entries, (3) one or more previously-defined parsing rule annotators, and (4) one or more characters that each represent a nucleotide;

creating a parsing rule annotator that identifies an order of and a combination of at least two elements selected from (1) the one or more lexical annotators, (2) the one or more dictionary entries, (3) the one or more previously-defined parsing rule annotators, and (4) the one or more characters that each represent a nucleotide; and

creating an Unstructured Information Management Architecture (UIMA) pipeline to locate the genome pattern using the parsing rule annotator by;

in a first stage of the UIMA pipeline, parsing a genetic sequence that is found in a Common Analysis Structure (CAS) to determine a language used and to generate tokens that are added to the CAS with a start position and an end position for each of the tokens;

in a second stage of the UIMA pipeline, executing the one or more lexical annotators against the genetic sequence to identify one or more lexical annotations that are added to the CAS with a start position and an end position for each of the one or more lexical annotations; and

in a third stage of the UIMA pipeline, using the start position and the end position for each of the tokens and the start position and the end position for each of the one or more lexical annotations to identify a match to the parsing rule annotation and to form a new annotation that is added to the CAS.

2. The method of claim 1, wherein the parsing rule annotator includes lexical annotators and no previously-defined parsing rule annotators.

3. The method of claim 1, wherein the parsing rule annotator includes previously-defined parsing rule annotators and no lexical annotators.

4. The method of claim 1, wherein a Software as a Service (SaaS) is provided to perform the method.

5. The method of claim 1, wherein there are multiple occurrences of a same lexical annotation with different start positions and end positions.

6. The method of claim 1, further comprising:

in response to determining that any token matches to a dictionary entry from the one or more dictionary entries, storing a new annotation in the CAS.

7. The method of claim 1, wherein each of the one or more dictionary entries represents a feature of a human.