Nonparametric inference for local extrema with application to oligonucleotide microarray data in yeast genome.

Area of Science:

• Bioinformatics
• Genomics
• Computational Biology

Background:

• Identifying local extrema in expression profiles is crucial for analyzing biological data, such as locating replication origins in yeast genomes.
• Current methods for detecting these extrema in microarray experiments have limitations.

Purpose of the Study:

• To develop and present a novel nonparametric kernel smoothing (NKS) technique for detecting local hybridization intensity extrema.
• To accurately identify putative replication origins and autonomous replication sequence (ARS) elements in the yeast genome.

Main Methods:

• The proposed method utilizes equilibrium points (where the first derivative of the intensity curve is zero) for inference.
• Nonparametric kernel smoothing (NKS) is applied to detect local extrema in hybridization intensity profiles.
• The technique provides both point and interval estimation for extrema locations and allows hypothesis testing.

Main Results:

• The NKS method successfully identified local extrema in yeast genome replication data.
• The technique located autonomous replication sequence (ARS) elements by analyzing equilibrium points.
• Several ARS elements were detected using NKS that were missed by conventional methods like Fourier convolution smoothing.

Conclusions:

• The proposed NKS technique offers a robust approach for identifying local extrema in expression profiles.
• This method enhances the detection of crucial genomic elements like yeast ARS elements.
• NKS provides a valuable tool for analyzing microarray data and advancing our understanding of genome replication dynamics.