What question did this study set out to answer?

To improve the accuracy of gene structure prediction in the Genie genefinding program by implementing neural network-based splice site sensors.

January 1, 1997

Improved splice site detection in Genie

View Full Paper

Ask AI

Bookmark

Authors

MRMartin G. ReeseID Genomics (United States)FEFrank H. EeckmanHuman Genome Sciences (United States)DKDavid KulpEmory University

Discussion

Member takes

Implication

Algorithmic evaluation demonstrates improved gene structure prediction accuracy in annotated DNA sequences, indicating neural networks enhance splice site identification.

Key Points

To improve the accuracy of gene structure prediction in the Genie genefinding program by implementing neural network-based splice site sensors.
Replaced existing splice site sensors within the generalized Hidden Markov Model (GHMM) framework of Genie with two novel neural networks based on dinucleotide frequencies.
Evaluated predictive performance using a standard benchmark dataset of annotated genes against the older sensor system, analyzing correlations with intron and exon lengths.
Identified 86% of coding nucleotides correctly with a specificity of 85%, compared to 80% sensitivity and 84% specificity in the prior system.
Demonstrated overall improvements in the sensitivity and specificity of full gene structure identification while assessing distance-dependent false positive rates.

Cite This Study

Reese et al. (1997) studied this question.

synapsesocial.com/papers/6a1e951fd258f1f64202aff6 https://doi.org/10.1145/267521.267766

View Full Paper

Ask AI

Bookmark

Implication

Algorithmic evaluation demonstrates improved gene structure prediction accuracy in annotated DNA sequences, indicating neural networks enhance splice site identification.

Improved splice site detection in Genie

Discussion

Member takes

Implication

Key Points

Cite This Study

Improved splice site detection in Genie

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider