New Genomics Technology: Copy Number Variation Analysis Methods Annelise Mah Genomics and Medicine Fall 2008 Final Paper 12/01/08 Professor Doug Brutlag Stanford UniversityWhat are Copy Number Variations? CNVs, or sometimes CNPs (Copy Number Polymorphisms), are changes in the number of appearances a certain pattern makes in the genome. Copy Number Variable Regions (CNVRs) are regions of the genome that are copied, deleted, or varied in number in some way. Normally these regions are defined as a kilobase (Kb, 10^3) to several megabases (Mb, 10^6) in size. These CNVRs make up around 12% of the human genome, cause disease, affect gene expression, and alter the organism’s phenotype. A total of 1447 CNVRs spanning 360 Mb and associated with over 3000 genes has been discovered. (1) So far, it has been estimated that human individuals differ from each other by anywhere from 4 to 20 Mb (2,3). CNVs have in the past thought to be much rarer; several researchers discovered their ubiquity within the last few years and, since then, many new studies have been conducted. This paper will cover several techniques that have been used to discover CNVs. Methods for Finding CNVs ROMA (Representational Oligonucleotide Microarray Analysis) ROMA was one of the first methods designed to randomly search for CNVs. For this method, scientists choose selective DNA fragments (representations) and design oligonucleotide (around 50-100 base pairs) probes for them. These are laid on a microarray with different wells containing many copies each probe. More recently, microarray chips are made, either by printing on glass or synthesizing onto silica by laser photochemistry, with many copies of a single probe comprising each dot. Two sets of genes are then tagged with different fluorescent dyes, usually green or red, following the common microarray method. Often, a reference gene and a tumor gene are used in order to compare the CNVs of the two. These genes are then hybridized to an array with those genes on it. If a dot glows yellow, then the gene is present equally in bothsamples. A mostly-red or a mostly-green dot will indicate a deletion or replication in the gene. Depending on the color’s intensity, an estimation of the number of copies/deletions can be made. A hidden markov model or other algorithm is used to accurately predict the difference in copy number. (4) One problem with the ROMA technique is that its probes do not cover very much of the genome. (Fanciful illustration of oligonucleotide probes with tagged DNA attached, Illumina) Fosmid Paired-End Sequencing In 2005, Tuzen et al. used a novel comparative method to locate deletions, insertions, and even inversions in the human genome that could denote CNVs. The group sequenced the ends of fosmids (bacteria vectors with inserted human DNA) containing a human genome. These ends were a known distance apart, around 40 kb. The sequences of these end-pairs were then compared to a reference genome. If the sequences that made up the ends fell much shorter or farther apart on the reference genome than predicted by the fosmid size, then an insertion or deletion must have occurred in the test genome. If the sequences were unexpectedly flipped, then an inversion could have occurred. They also checked the insertions they found by designing PCR primers at both ends and the middle of the change. If the shorter segments that started in the inserted DNA showed up as well as the whole sequence, then DNA was inserted. If only theentire sequence was copied, then no DNA was inserted. (5) This method is now considered limited in scope, but it showed around 300 CNVRs, including many that were related to disease. SKY (Spectral Karyotyping) or FISH (Fluorescence In Situ Hybridization) In situ, or in place, means that the reaction takes place between in vivo and in vitro, in its natural tissue but not in the organism (Merriam-Webster, Wikipedia). To detect CNVs, probes of target genes are fluorescently tagged and hybridized to chromosome or sections of DNA to see where they land. (6) These methods are not that accurate discovery or quantification tools, and are mostly used to validate finds and map them to a certain location. (SKY coloring of chromosomes, Medical University of South Carolina) (CNV seen using FISH) CGH (Comparative Genomic Hybridization) For a general look at CNV, scientists can isolate metaphase chromosomes and hybridize both tagged tumor genes and tagged normal genes to the normal chromosome. Different areas of the chromosome will glow different colors based on the variations in copy number in the tumor and reference gene. However, this strategy gives a global view, and really only operates at the Mb level (7). It is considered too low-resolution, unable to detect smaller changes.RT/Q-PCR (Real Time Quantitative PCR) RT-PCR or Q-PCR is another strategy that is mainly used to confirm or validate finds. PCR, Polymerase Chain Reaction, is a technique used to quickly amplify a sample of DNA. The elements of PCR are fluorescently tagged and the intensity of the glow is monitored between cycles. Comparatively, a single gene with more copies will increase sooner than one with fewer copies when the fluorescent glow is measured on a logarithmic scale. A more dilute sample will increase later (graphically, shifted to the right). This technique is not perfect because of uncontrollable environmental factors, but it can give a comparative quantification of copy numbers (8). (Logarithmic graph of fluorescent intensity by time, 8) aCGH (Array Comparative Genomic Hybridization) Array Comparative Genomic Hybridization (CGH) has become one of two new powerful array-based methods. Rather than picking certain representative probes, scientists using aCGH can use most of the genome. Genes of interest are inserted into a BAC library (Bacterial Artificial Chromosome, usually e. coli holding human DNA), for concentration and replication. Normal microarray procedure is followed—these genes are spliced into fragments, labeled withfluorescent tags, and hybridized onto a microarray containing the 1genes of interest. Based on the color of each dot, the color’s intensity, and a complicated algorithm, the amount of copying or deletion can be estimated. (9) The data is then “smoothed” so that the gene is deleted or copied in integer fashion. These models show areas of deletion or copying, as well as “breakpoints” in the genome. These breakpoints are
View Full Document
Unlocking...