Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2008 Feb 6;3(2):e1546.
doi: 10.1371/journal.pone.0001546.

Systematic validation and atomic force microscopy of non-covalent short oligonucleotide barcode microarrays

Affiliations

Systematic validation and atomic force microscopy of non-covalent short oligonucleotide barcode microarrays

Michael A Cook et al. PLoS One. .

Abstract

Background: Molecular barcode arrays provide a powerful means to analyze cellular phenotypes in parallel through detection of short (20-60 base) unique sequence tags, or "barcodes", associated with each strain or clone in a collection. However, costs of current methods for microarray construction, whether by in situ oligonucleotide synthesis or ex situ coupling of modified oligonucleotides to the slide surface are often prohibitive to large-scale analyses.

Methodology/principal findings: Here we demonstrate that unmodified 20mer oligonucleotide probes printed on conventional surfaces show comparable hybridization signals to covalently linked 5'-amino-modified probes. As a test case, we undertook systematic cell size analysis of the budding yeast Saccharomyces cerevisiae genome-wide deletion collection by size separation of the deletion pool followed by determination of strain abundance in size fractions by barcode arrays. We demonstrate that the properties of a 13K unique feature spotted 20 mer oligonucleotide barcode microarray compare favorably with an analogous covalently-linked oligonucleotide array. Further, cell size profiles obtained with the size selection/barcode array approach recapitulate previous cell size measurements of individual deletion strains. Finally, through atomic force microscopy (AFM), we characterize the mechanism of hybridization to unmodified barcode probes on the slide surface.

Conclusions/significance: These studies push the lower limit of probe size in genome-scale unmodified oligonucleotide microarray construction and demonstrate a versatile, cost-effective and reliable method for molecular barcode analysis.

PubMed Disclaimer

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.

Figures

Figure 1
Figure 1. Sensitivity and specificity of spotted 20mer barcode oligonucleotides.
A. Fluorescent images of hybridized pilot barcode arrays fabricated on Superaldehyde® (top; substrate permissive to covalent linkage of probes) and GAPS™II (bottom; substrate non-permissive to covalent linkage of probes) slides. 184 barcode oligonucleotides were printed as two sets of quadruplet features in each of two identical arrays (or “super-grids”; only the top super-grid is shown for each substrate). Pairs of amino-modified (white circles) and unmodified control probes (white squares) are shown linked by dotted lines. A defective modified barcode, excluded from further analysis, is noted (white asters). Negative controls are shown (white dashed box). All other unmarked features hybridized with barcodes from sub-populations of non-essential deletion strains of chromosome 2, labeled with either Cy5 (Chr2_1) or Cy3 (Chr2_2). B. Effect of 5′-amino-modification on oligonucleotide performance. Performance of each of seven different barcode oligonucleotide pairs (sequences are provided in a Supplemental GeneList file; List Data S1; the probe pairs correspond to YBL090W-DN, YBL091C-DN, YBL091C-UP, YBL093C-DN, YBL093C-UP, YBL094C-DN, YBL094C-UP, from left to right) is measured by log2 ratio of SNR(unmodified)/SNR(modified) on Superaldehyde® substrate, corrected by subtracting the log2 ratio of that on GAPS™II substrate; where SNR denotes signal to noise ratio of the probe [(median intensity Cy5 [or Cy3] - median background Cy5 [or Cy3])/standard deviation background Cy5 (or Cy3)]. Performance was determined based on hybridization to the cognate barcode sequences (barcode, from Figure 1A) or to a Cy3-labeled randomized 9mer probe. C. Specificity and signal intensity of hybridized barcodes in the absence of covalent linkage to substrate. The log2 ratio of the background subtracted Cy5/Cy3 channels (M) is plotted versus the average log2 value of the signal intensity in each channel (A) for 8 replicates of each barcode. Barcodes from sub-populations of non-essential deletion strains from chromosome 2 are labeled with Cy5 (Chr2_1; black [amino-modified] and gold [unmodified]) and Cy3 (Chr2_2; magenta). Negative control sequences are shown (red). Spots with background-subtracted intensities below zero for one channel fall along straight lines. Negative values were assigned an arbitrary floor value of 1 (log2 = 0) in calculations of A and M. (The fluorescent image of 4 replicates for each probe is shown in the bottom panel of Figure 1A.)
Figure 2
Figure 2. Features of the 13K unique feature SUBarray.
A. A representative microarray from an experiment with the haploid yeast deletion set. Arrays are constructed of 48 blocks. B. Enlargement of a region overlapping two of the blocks. In white boxes are the four oligonucleotide controls that are present within every block. C. Logarithmic scale scatter plot of background subtracted intensity for Cy5 versus Cy3-dye for a representative barcode elutriation cell size experiment. Barcodes of strains not present within the experimental pool (black) and the four sets of positive controls (magenta) are overlayed on the barcodes of strains represented in the pool (blue). D. Overlap of strains represented by at least one significant barcode signal between Agilent covalently linked 20mer arrays and SUBarrays. Significant barcode signals are as defined in the text.
Figure 3
Figure 3. Barcode elutriation cell size experiment.
A. Schematic of the experiment. Elutriation of the haploid pool enriches for small cells. Genomic DNA is isolated from cell populations immediately before and immediately after elutriation, differentially labeled with either Cy5 or Cy3, and applied to a barcode microarray. B. Cell size distributions of a characteristic elutriation as determined by use of a Coulter Z2 particle analyzer (Beckman). Increasing the rate of flow through the elutriation rotor increases the median cell size of the elutriated fraction. C. Scatter plot of the average barcode Z score. Agilent (Y axis). SUBarray (X axis). Left. Overlay of barcodes of strains confirmed as either lge (black), whi (magenta), or wild-type (WT; red) by systematic experiments on all other barcodes (blue). Right. Overlay of barcodes of structural components of the cytosolic or mitochondrial ribosome (magenta) or respiration defective strains (black) on all other barcodes (blue). D. Venn diagrams of overlap between systematic and barcode data using SUBarrays or Agilent arrays. Cell size mutants are as defined in the text. E. Overlap between systematic and barcode data. Clustered cell size distributions are represented by horizontal bars color coded by intensity to reflect the shape of the distributions, with deletion strains on the vertical, and cell size on the horizontal axis. Barcodes, divided into UP (U) and DN (D) tags, with Z scores greater than 1 are represented as Lge (green bars) or Whi (red bars) adjacent to their cognate deletion strain for all experiments (E1, E2, E3) and all elutriation cuts (16, 21, 24 mL/min). SUBarrays and Agilent arrays are shown. The density of lge or whi barcodes (Z score>1) over a 3 gene window is represented by horizontal bars color coded by intensity (>10% brown; >30% yellow; >70% orange). Only barcodes with significant intensity/SNR were used in density calculations. Gene deletion strains with systematic mean cell sizes >1 SD larger (green) or smaller (red) than average are shown. Systematic lge strains with broad distributions and unexpected cell size by barcode are indicated (dotted red box and asterisk).
Figure 4
Figure 4. Barcode elutriation and systematic cell size enrichment of gene ontology (GO) annotations.
Statistical enrichment from the haploid deletion set was calculated using cumulative hypergeometric probability functions (CDFs) with Bonferroni correction for either the high confidence systematic cell size data; SUBarray or Agilent array data; or the combined set of all whi or lge deletion strains identified. Significant enrichment for whi (red bars) or lge (green bars) strains is defined by [p*N<0.01, where N is the number of categories per subset)]. GO component (N = 330); GO process (N = 1380); Morphology (N = 11) . GO categories are arranged by related function.
Figure 5
Figure 5. Atomic force microscopy (AFM) of hybridized and mock hybridized spotted 20mer barcodes.
A. Fluorescent imaging of hybridized and mock hybridized slides. Each slide contained two spots of oligonucleotides complementary or non-complementary to the labeled target sequence. The location of the AFM scans is indicated by a red arrow. B. Mock hybridized barcode oligonucleotide. 1 µm×1 µm scan of the surface arrangement. Pixels are ∼2 nm on the x and y axis, and <100 pm on the z axis. C. High density non-complementary barcode oligonucleotide hybridized in the presence of its non-cognate target sequence. The location of a region of aberrant signal omitted from the analysis is indicated by a red arrow. D. Complementary barcode oligonucleotide hybridized in the presence of its cognate target sequence. E. Cumulative frequency of peak number with increasing height cut-off for high (solid blue) and low density (dashed blue) non-complementary and complementary barcode oligonucleotides minus (green) or plus (red) hybridized target. F. Cumulative frequency of peak number with increasing diameter cut-off for high density (solid blue) and low density (dashed blue) non-complementary and complementary barcode oligonucleotides minus (green) or plus (red) hybridized target. The method of estimation of peak dimensions is described in the Supplementary material (Supplemental text S1). The height and diameter at which 99% of mock hybridized peaks are excluded is indicated by a red dotted line.

References

    1. Glynn EF, Megee PC, Yu HG, Mistrot C, Unal E, et al. Genome-wide mapping of the cohesin complex in the yeast Saccharomyces cerevisiae. PLoS Biol. 2004;2:E259. - PMC - PubMed
    1. Hoheisel JD. Microarray technology: beyond transcript profiling and genotype analysis. Nat Rev Genet. 2006;7:200–210. - PubMed
    1. Raghuraman MK, Winzeler EA, Collingwood D, Hunt S, Wodicka L, et al. Replication dynamics of the yeast genome. Science. 2001;294:115–121. - PubMed
    1. Shivaswamy S, Iyer VR. Genome-wide analysis of chromatin status using tiling microarrays. Methods. 2007;41:304–311. - PMC - PubMed
    1. Wu J, Smith LT, Plass C, Huang TH. ChIP-chip comes of age for genome-wide functional analysis. Cancer Res. 2006;66:6899–6902. - PubMed

Publication types

MeSH terms