Transcript Quantitative analysis to assess the performance of the

Quantitative analysis to assess the performance of the complete
Agilent oligo aCGH microarray system.
Sunny Song, Peter Webb, Anniek De Witte, Diane Ilsley and Jim Collins
Agilent Technologies, Santa Clara, CA 95051
ABSTRACT
Comparative genomic hybridization (CGH) is a technique for studying
chromosomal changes in cancer. As cancerous cells multiply, they can undergo
dramatic chromosomal changes, including chromosome loss, duplication, and
the translocation of DNA from one chromosome to another. Chromosome
aberrations have previously been detected using optical imaging of whole
chromosomes, a technique with limited sensitivity, resolution, quantification, and
throughput. Efforts in recent years to use microarrays to overcome these
limitations have been hampered by inadequate sensitivity, specificity and
flexibility of the microarray systems.
The Agilent oligonucleotide CGH microarray system overcomes several
scientific hurdles that have impeded comparative genomic studies of cancer.
This new system can reliably detect single copy deletions in chromosomes. The
system includes a whole human genome microarray, reagents for sample
preparation, an optimized microarray processing protocol, and software for data
analysis and visualization.
In this study, we demonstrate the ability of the Agilent oligo 244K CGH system to
reliably detect DNA copy variations with reduced sample input amounts. Further,
we illustrate the benefits of high resolution CGH in mapping chromosomal
breakpoints in cancer cells.
Fig. 4 244K microarray platform performance using a range of genomic DNA
input masses. A) Log Ratio Noise. The standard deviation of Log2 Ratio values is < 0.25
across the range of input masses tested. B) Reproducibility across replicate probes.
The %CV values for intra-array replicate probes are all less than < 5 % for the different DNA input
masses. C) Accuracy. The average Log2 Ratio of chrX probes in Male/Female hybridizations is
greater than 0.9 across the DNA input masses tested indicating a high level of accuracy.
EXPERIMENTAL OVERVIEW
Fig. 2. Probe coverage on the 244K microarray. Typical
• The Human Genome CGH 244K microarray (p/n G4411B) and the Human
Genome CGH 44K (p/n G4410B) were used, which contained In situ synthesized
60 mer oligo probes covering both the coding and non-coding sequences of total
genomic DNA with an average spatial resolution of approximately 6 kb and 35 kb,
respectively.
probe coverage characteristics on the Human Genome CGH 244K
and 44K microarrays illustrated using the UCSC hg18 Human
Genome Browser, Feb 2006 (NCBI build 36) of human chromosome
17. At the top of each view, each short vertical bar represents a
single probe on Agilent Human Genome CGH 244K (red) or 44K
(blue) microarray. Annotations below the probes indicate the location
mapping of UCSC known genes (blue or black) and Sanger
microRNAs (red). A) Variable probe density. A 3.0 MB
microarray window of 17q24-25 shows comprehensive coverage that
is more dense in coding regions. B) Coverage of long genes.
A 3.0 MB window of 17q11-12 depicts evenly distributed probe
coverage across long (ACCN1) as well as short (NLE1) genes. C)
Coverage of miRNAs. A 1.0 MB region of 17q11.2 illustrates
microRNA representation (hsa-mir-193, hsa-mir-365-2).
• Total genomic DNA was used for direct labeling. Human genomic DNA from
pooled Male, Female, and colon carcinoma cell line HT-29 were labeled using the
Agilent Genomic DNA Labeling Kit PLUS (p/n 5188-5309). Each reaction started
with 0.2 mg, 0.5 mg, or 3 mg of total genomic DNA input, and the entire reaction
product was hybridized onto the microarray using the Agilent aCGH Hybridization
Kit (p/n 5188-5220) using Agilent standard CGH protocol (version 4.0). The data
in each graph represents an average of two or more microarrays.
• Microarrays were washed using the Agilent aCGH Wash Kit (p/n 5188-5226), and
scanned using the Agilent dual laser DNA microarray scanner (p/n G2565BA).
Data were extracted using Agilent’s Feature Extraction version 9.1 software. Data
were analyzed using Agilent’s CGH Analytics version 3.4 software.
Fig. 3. Probe Performance on the 244K microarray.
A) Low noise and high specificity. aCGH 244K
Fig. 1. Agilent’s Microarray Comparative Genomic
Hybridization (CGH) Whole Platform Solution. Schematic depiction
of the procedures used to measure DNA copy-number changes by CGH
microarray hybridization. Genomic DNA samples isolated from tumor cells and
normal cells are labeled with two different fluorophores (Cy5 and Cy3,
respectively) and hybridized together to a single CGH microarray. For each
DNA probe on the microarray, the ratio of intensity of the fluorescence
measured for the two fluors is determined by dual laser scanning and data are
analyzed using the CGH Analytics software. A red color represents increased
DNA copy number, green represents decreased copy number (i.e., deletion),
and yellow represents no change in DNA copy number in tumor cell DNA
compared with normal cell DNA.
microarray log2 ratios are shown for male/female hybridizations. A
representative separation histogram showing the distribution of
red/green fluorescence ratios for the autosomal probes (green)
and X-chromosomal probes (blue), plotted as fraction of probes
on the ordinate (female, XX) versus red/green fluorescence ratio
(binned by intervals) on the abscissa of the X-chromosome (male,
XY) on the Agilent Human Genome CGH 244K microarray. B)
Improved signal intensity. Signal intensities of probes on
the 244K microarray show a tighter distribution and medians of
log10 signal Intensities. 244K microarray (red) and 44K
microarray (blue) were hybridized with male/female DNA. More
probes on the 244K microarray show higher signal intensities
while the distribution of these intensities is tighter.
Fig. 5 The high resolution
244K microarray provides
enhanced precision that
enables novel aberration
detection . CGH Analytics views
of Agilent 244K and 44K analyses of
chromosome 3 (A, B) and
chromosome 16 (C, D) in the human
colon carcinoma cell line HT29. A)
Left, Scatter plot (chromosome
view) produced from Human CGH
Microarray 244K analysis reveals
multiple 3p arm deletions in regions
such as 3p12.1-12.3 (horizontal shift
to left of zero line) and a focal
deletion in 3p14.2 (horizontal shift to
left of zero line, outlined by dotted
blue box). Right, Zoomed-in gene
view which focuses on a 7 MB
window within 3p14.2 containing the
focal deletion. It distinctly shows
two different deletion patterns
(homozygous and heterozygous
deletion) both within the single
tumor suppressor FHIT gene. B)
Parallel scatter plots (chromosome
and gene views, respectively) from
CGH 44K Microarray analysis of the
same sample as in panel A. C) Left,
Scatter plot (chromosome view)
produced from 244K analysis
reveals a focal homozygous deletion
in 16p13.2. Right, Zoomed-in gene
view which focuses on a 0.5 MB
window within 16p13.2 containing a
homozygous deletion. It readily
detects multiple deletion patterns
within the ataxin-2 binding protein
A2BP1 gene (green dots).
Further, it shows a ~ 50KB microdeletion within 16q23.1 (green dots in red circle). This
microdeletion (heterozygous deletion) was verified by multiple consecutive probes. D)
Parallel scatter plots from CGH 44K Microarray analysis of the same sample as in panel C.
CONCLUSIONS
• Agilent’s 244K CGH microarray platform allows researchers to identify DNA
copy variations with high resolution on a genome-wide scale.
• High quality data can be obtained from low input masses of genomic DNA
without the need for sample amplification or complexity reduction.
• This high density platform provides enhanced precision that enables novel
aberration detection.