MetaProm: A neural network based meta-predictor for alternative human promoter prediction

Junwen Wang, Lyle H. Ungar, Hung Tseng, Sridhar Hannenhalli

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

Background: De novo eukaryotic promoter prediction is important for discovering novel genes and understanding gene regulation. In spite of the great advances made in the past decade, recent studies revealed that the overall performances of the current promoter prediction programs (PPPs) are still poor, and predictions made by individual PPPs do not overlap each other. Furthermore, most PPPs are trained and tested on the most-upstream promoters; their performances on alternative promoters have not been assessed. Results: In this paper, we evaluate the performances of current major promoter prediction programs (i.e., PSPA, FirstEF, McPromoter, DragonGSF, DragonPF, and FProm) using 42,536 distinct human gene promoters on a genome-wide scale, and with emphasis on alternative promoters. We describe an artificial neural network (ANN) based meta-predictor program that integrates predictions from the current PPPs and the predicted promoters' relation to CpG islands. Our specific analysis of recently discovered alternative promoters reveals that although only 41% of the 3′ most promoters overlap a CpG island, 74% of 5′ most promoters overlap a CpG island. Conclusion: Our assessment of six PPPs on 1.06 × 109 bps of human genome sequence reveals the specific strengths and weaknesses of individual PPPs. Our meta-predictor outperforms any individual PPP in sensitivity and specificity. Furthermore, we discovered that the 5′ alternative promoters are more likely to be associated with a CpG island.

Original languageEnglish (US)
Article number374
JournalBMC Genomics
Volume8
DOIs
StatePublished - Oct 17 2007
Externally publishedYes

Fingerprint

CpG Islands
Genes
Human Genome
Genome
Sensitivity and Specificity

ASJC Scopus subject areas

  • Medicine(all)

Cite this

MetaProm : A neural network based meta-predictor for alternative human promoter prediction. / Wang, Junwen; Ungar, Lyle H.; Tseng, Hung; Hannenhalli, Sridhar.

In: BMC Genomics, Vol. 8, 374, 17.10.2007.

Research output: Contribution to journalArticle

Wang, Junwen ; Ungar, Lyle H. ; Tseng, Hung ; Hannenhalli, Sridhar. / MetaProm : A neural network based meta-predictor for alternative human promoter prediction. In: BMC Genomics. 2007 ; Vol. 8.
@article{da45dc31c3f24c28975ed90f104740b8,
title = "MetaProm: A neural network based meta-predictor for alternative human promoter prediction",
abstract = "Background: De novo eukaryotic promoter prediction is important for discovering novel genes and understanding gene regulation. In spite of the great advances made in the past decade, recent studies revealed that the overall performances of the current promoter prediction programs (PPPs) are still poor, and predictions made by individual PPPs do not overlap each other. Furthermore, most PPPs are trained and tested on the most-upstream promoters; their performances on alternative promoters have not been assessed. Results: In this paper, we evaluate the performances of current major promoter prediction programs (i.e., PSPA, FirstEF, McPromoter, DragonGSF, DragonPF, and FProm) using 42,536 distinct human gene promoters on a genome-wide scale, and with emphasis on alternative promoters. We describe an artificial neural network (ANN) based meta-predictor program that integrates predictions from the current PPPs and the predicted promoters' relation to CpG islands. Our specific analysis of recently discovered alternative promoters reveals that although only 41{\%} of the 3′ most promoters overlap a CpG island, 74{\%} of 5′ most promoters overlap a CpG island. Conclusion: Our assessment of six PPPs on 1.06 × 109 bps of human genome sequence reveals the specific strengths and weaknesses of individual PPPs. Our meta-predictor outperforms any individual PPP in sensitivity and specificity. Furthermore, we discovered that the 5′ alternative promoters are more likely to be associated with a CpG island.",
author = "Junwen Wang and Ungar, {Lyle H.} and Hung Tseng and Sridhar Hannenhalli",
year = "2007",
month = "10",
day = "17",
doi = "10.1186/1471-2164-8-374",
language = "English (US)",
volume = "8",
journal = "BMC Genomics",
issn = "1471-2164",
publisher = "BioMed Central",

}

TY - JOUR

T1 - MetaProm

T2 - A neural network based meta-predictor for alternative human promoter prediction

AU - Wang, Junwen

AU - Ungar, Lyle H.

AU - Tseng, Hung

AU - Hannenhalli, Sridhar

PY - 2007/10/17

Y1 - 2007/10/17

N2 - Background: De novo eukaryotic promoter prediction is important for discovering novel genes and understanding gene regulation. In spite of the great advances made in the past decade, recent studies revealed that the overall performances of the current promoter prediction programs (PPPs) are still poor, and predictions made by individual PPPs do not overlap each other. Furthermore, most PPPs are trained and tested on the most-upstream promoters; their performances on alternative promoters have not been assessed. Results: In this paper, we evaluate the performances of current major promoter prediction programs (i.e., PSPA, FirstEF, McPromoter, DragonGSF, DragonPF, and FProm) using 42,536 distinct human gene promoters on a genome-wide scale, and with emphasis on alternative promoters. We describe an artificial neural network (ANN) based meta-predictor program that integrates predictions from the current PPPs and the predicted promoters' relation to CpG islands. Our specific analysis of recently discovered alternative promoters reveals that although only 41% of the 3′ most promoters overlap a CpG island, 74% of 5′ most promoters overlap a CpG island. Conclusion: Our assessment of six PPPs on 1.06 × 109 bps of human genome sequence reveals the specific strengths and weaknesses of individual PPPs. Our meta-predictor outperforms any individual PPP in sensitivity and specificity. Furthermore, we discovered that the 5′ alternative promoters are more likely to be associated with a CpG island.

AB - Background: De novo eukaryotic promoter prediction is important for discovering novel genes and understanding gene regulation. In spite of the great advances made in the past decade, recent studies revealed that the overall performances of the current promoter prediction programs (PPPs) are still poor, and predictions made by individual PPPs do not overlap each other. Furthermore, most PPPs are trained and tested on the most-upstream promoters; their performances on alternative promoters have not been assessed. Results: In this paper, we evaluate the performances of current major promoter prediction programs (i.e., PSPA, FirstEF, McPromoter, DragonGSF, DragonPF, and FProm) using 42,536 distinct human gene promoters on a genome-wide scale, and with emphasis on alternative promoters. We describe an artificial neural network (ANN) based meta-predictor program that integrates predictions from the current PPPs and the predicted promoters' relation to CpG islands. Our specific analysis of recently discovered alternative promoters reveals that although only 41% of the 3′ most promoters overlap a CpG island, 74% of 5′ most promoters overlap a CpG island. Conclusion: Our assessment of six PPPs on 1.06 × 109 bps of human genome sequence reveals the specific strengths and weaknesses of individual PPPs. Our meta-predictor outperforms any individual PPP in sensitivity and specificity. Furthermore, we discovered that the 5′ alternative promoters are more likely to be associated with a CpG island.

UR - http://www.scopus.com/inward/record.url?scp=38049156077&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=38049156077&partnerID=8YFLogxK

U2 - 10.1186/1471-2164-8-374

DO - 10.1186/1471-2164-8-374

M3 - Article

C2 - 17941982

AN - SCOPUS:38049156077

VL - 8

JO - BMC Genomics

JF - BMC Genomics

SN - 1471-2164

M1 - 374

ER -