|
Advances in DNA microarray fabrication technologies, expanding probe libraries, and new bioinformatics methods and resources have firmly established array-based techniques as mainstream bioanalytical tools and the application space is proliferating rapidly. However, the capability of these tools to yield truly quantitative information remains limited, primarily due to problems inherent to the use of fluorescence imaging for reading the hybridized arrays. The obvious advantages of fluorescence are the unrivaled sensitivity and simplicity of the instrumentation. There are disadvantages of this approach, however, such as difficulties in achieving optimal labeling of targets and reproducible signals (due to quenching, resonance energy transfer, photobleaching effects, etc.) that undermine precision. We are exploring alternative approaches, based mainly on Raman and resonance Raman spectroscopy, that in principle permit direct analysis of structural differences between hybridized and unhybridized probes, thereby eliminating the need for labeling the target analytes. We report here on the status of efforts to evaluate the potential of these methods based on a combination of measured data and simulated experiments involving short (12-mer) ssDNA oligomer probes with varying degrees of hybridized target DNA. Preliminary results suggest that it may be possible to determine the fraction of duplex probes within a single register on a DNA microarray from 100% down to 10% (or possibly less) with a precision of ±2 5%. Details of the methods used, their implementation, and their potential advantages and limitations are presented, along with discussion of the utility of using 2DCOS methods to emphasize small spectral changes sensitive to interstrand H bonding, backbone flexibility, hypochromicity due to base-stacking in duplex structures and solvation effects.
|
