Polony sequencing : DNA sequencing technology and a computational analysis reveals chromosomal domains of gene expression

Abstract

The first part of this thesis describes the development of polony sequencing, a sequencing technology in which DNA is cloned, amplified and sequenced in a polymer matrix. A complex library of one to ten million linear DNA molecules is amplified by performing polymerase chain reaction (PCR) in a thin polyacrylamide film poured on a glass microscope slide. The polyacrylamide matrix retards the diffusion of the DNA molecules so that each amplification product remains localized near its parent molecule. At the end of the reaction, a number of polymerase colonies, or "polonies", have formed, each one grown from a single template molecule. As many as 5 million clones can be amplified in parallel on a single slide. By including an acrydite modification at the 5' end of one of the PCR primers, the amplified DNA will be covalently attached to the polyacrylamide matrix, allowing further enzymatic manipulations to be performed on all clones simultaneously. Also described in this thesis is my progress in development of a protocol to sequence the polonies by repeated cycles of extension with fluorescent deoxynucleotide. Because polony sequencing is inherently parallel, and sub-picoliter volumes are used for each reaction, the technology should be substantially faster and cheaper than existing methods. Applications for polony sequencing such as gene expression analysis, SNP discovery, and SNP screening will also be discussed. The second part of this thesis describes a computational analysis that tests the hypothesis that chromosomal position affects gene expression. It is shown that, throughout the genome, genes lying close together on the same chromosome often show significant coexpression. This coexpression is independent of the orientation of genes to each other, but is dependent on the distance between genes. In several cases where adjacent genes show highly correlated expression, the promoter of only one of the genes contains an upstream activating sequence (UAS) known to be associated with the expression pattern. These results suggest that in certain regions of the genome a single transcription factor binding site may regulate several genes. It is also shown that evolution may take advantage of this phenomenon by keeping genes with similar functions in adjacent positions along the chromosomes. The techniques that are presented provide a computational method to delineate the locations of chromosomal domains and identify the boundary elements that flank them.