Document Detail


Splice site prediction using stochastic regular grammars.
MedLine Citation:
PMID:  17469059     Owner:  NLM     Status:  MEDLINE    
Abstract/OtherAbstract:
This paper presents a novel approach to the problem of splice site prediction, by applying stochastic grammar inference. We used four grammar inference algorithms to infer 1465 grammars, and used 10-fold cross-validation to select the best grammar for each algorithm. The corresponding grammars were embedded into a classifier and used to run splice site prediction and compare the results with those of NNSPLICE, the predictor used by the Genie gene finder. We indicate possible paths to improve this performance by using Sakakibara's windowing technique to find probability thresholds that will lower false-positive predictions.
Authors:
A Y Kashiwabara; D C G Vieira; A Machado-Lima; A M Durham
Related Documents :
12144189 - Transparenchymal retraction maneuver: a simple technique in laparoscopic hepatectomy.
9050479 - Novel technique and instrumentation for laparoscopic application of hemostatic clips.
8429649 - Differential lung ventilation. applications beyond the operating room.
12633629 - Snake bite in australia: first aid and envenomation management.
17491459 - Accelerating the optimization of requantization codebook for vector quantization.
22431199 - Hearing aid processing changes tone burst onset: effect on cortical auditory evoked pot...
12853869 - Designing effective hiv/aids social and behavioral interventions for the population of ...
19477739 - Medical treatment for men who have sex with men and are living with hiv/aids.
10288859 - A strategic business issue.
Publication Detail:
Type:  Journal Article; Research Support, Non-U.S. Gov't; Validation Studies     Date:  2007-03-20
Journal Detail:
Title:  Genetics and molecular research : GMR     Volume:  6     ISSN:  1676-5680     ISO Abbreviation:  Genet. Mol. Res.     Publication Date:  2007  
Date Detail:
Created Date:  2007-04-30     Completed Date:  2007-08-01     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  101169387     Medline TA:  Genet Mol Res     Country:  Brazil    
Other Details:
Languages:  eng     Pagination:  105-15     Citation Subset:  IM    
Affiliation:
Departamento de Ciência da Computação, Instituto de Matemática e Estatística, Universidade de São Paulo, São Paulo, SP, Brasil.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Algorithms*
Artificial Intelligence*
Humans
Models, Molecular*
RNA Splicing / genetics*
Stochastic Processes

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  The use of the acridine orange test and the TUNEL assay to assess the integrity of freeze-dried bovi...
Next Document:  Cassava diversity in Brazil: the case of carotenoid-rich landraces.