Limits...
A new systematic computational approach to predicting target genes of transcription factors.

Dai X, He J, Zhao X - Nucleic Acids Res. (2007)

Bottom Line: Utilizing gene co-expression data, we modeled the prediction problem as a 'yes' or 'no' classification task by converting biological sequences into novel reverse-complementary position-sensitive n-gram profiles and implemented the classifiers with support vector machines.Our approach does not necessarily predict new DNA binding sites, which other studies have shown to be difficult and inaccurate.We applied the proposed approach to predict auxin-response factor target genes from published Arabidopsis thaliana co-expression data and obtained satisfactory results.

View Article: PubMed Central - PubMed

Affiliation: Plant Biology Division, the Samuel Robert Noble Foundation, Ardmore, OK 73401, USA.

ABSTRACT
Identifying transcription factor target genes (TFTGs) is a vital step towards understanding regulatory mechanisms of gene expression. Methods for the de novo identification of TFTGs are generally based on screening for novel DNA binding sites. However, experimental screening of new binding sites is a technically challenging, laborious and time-consuming task, while computational methods still lack accuracy. We propose a novel systematic computational approach for predicting TFTGs directly on a genome scale. Utilizing gene co-expression data, we modeled the prediction problem as a 'yes' or 'no' classification task by converting biological sequences into novel reverse-complementary position-sensitive n-gram profiles and implemented the classifiers with support vector machines. Our approach does not necessarily predict new DNA binding sites, which other studies have shown to be difficult and inaccurate. We applied the proposed approach to predict auxin-response factor target genes from published Arabidopsis thaliana co-expression data and obtained satisfactory results. Using ten-fold cross validations, the area under curve value of the receiver operating characteristic reaches around 0.73.

Show MeSH

Related in: MedlinePlus

Vector representation of DNA sequence using the featured reverse-complementary position sensitive n-grams.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC1935008&req=5

Figure 2: Vector representation of DNA sequence using the featured reverse-complementary position sensitive n-grams.

Mentions: A higher IG value indicates greater information significance, and thus suggests that the corresponding n-gram is better able to represent an important feature of the sequence. Here, we chose the top K (K = 500, 1000, 1500 and 2000) n-grams to represent the features of the DNA sequences, thereby constructing a K-dimensional vector space. The upstream DNA sequences were then converted into the K-dimensional vector space according to their n-gram profiles. Figure 2 is an example showing the conversion of a sequence into vector format in terms of these featured n-grams.Figure 2.


A new systematic computational approach to predicting target genes of transcription factors.

Dai X, He J, Zhao X - Nucleic Acids Res. (2007)

Vector representation of DNA sequence using the featured reverse-complementary position sensitive n-grams.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC1935008&req=5

Figure 2: Vector representation of DNA sequence using the featured reverse-complementary position sensitive n-grams.
Mentions: A higher IG value indicates greater information significance, and thus suggests that the corresponding n-gram is better able to represent an important feature of the sequence. Here, we chose the top K (K = 500, 1000, 1500 and 2000) n-grams to represent the features of the DNA sequences, thereby constructing a K-dimensional vector space. The upstream DNA sequences were then converted into the K-dimensional vector space according to their n-gram profiles. Figure 2 is an example showing the conversion of a sequence into vector format in terms of these featured n-grams.Figure 2.

Bottom Line: Utilizing gene co-expression data, we modeled the prediction problem as a 'yes' or 'no' classification task by converting biological sequences into novel reverse-complementary position-sensitive n-gram profiles and implemented the classifiers with support vector machines.Our approach does not necessarily predict new DNA binding sites, which other studies have shown to be difficult and inaccurate.We applied the proposed approach to predict auxin-response factor target genes from published Arabidopsis thaliana co-expression data and obtained satisfactory results.

View Article: PubMed Central - PubMed

Affiliation: Plant Biology Division, the Samuel Robert Noble Foundation, Ardmore, OK 73401, USA.

ABSTRACT
Identifying transcription factor target genes (TFTGs) is a vital step towards understanding regulatory mechanisms of gene expression. Methods for the de novo identification of TFTGs are generally based on screening for novel DNA binding sites. However, experimental screening of new binding sites is a technically challenging, laborious and time-consuming task, while computational methods still lack accuracy. We propose a novel systematic computational approach for predicting TFTGs directly on a genome scale. Utilizing gene co-expression data, we modeled the prediction problem as a 'yes' or 'no' classification task by converting biological sequences into novel reverse-complementary position-sensitive n-gram profiles and implemented the classifiers with support vector machines. Our approach does not necessarily predict new DNA binding sites, which other studies have shown to be difficult and inaccurate. We applied the proposed approach to predict auxin-response factor target genes from published Arabidopsis thaliana co-expression data and obtained satisfactory results. Using ten-fold cross validations, the area under curve value of the receiver operating characteristic reaches around 0.73.

Show MeSH
Related in: MedlinePlus