A Canvas of Spatially Arranged DNA Strands that Can Produce 24-bit Color Depth.

Nucleic acid microarray photolithography combines density, throughput, and positional control in DNA synthesis. These surface-bound sequence libraries are conventionally used in large-scale hybridization assays against fluorescently labeled, perfect-match DNA strands. Here, we introduce another layer of control for in situ microarray synthesis─hybridization affinity─to precisely modulate fluorescence intensity upon duplex formation. Using a combination of Cy3-, Cy5-, and fluorescein-labeled targets and an ensemble of truncated DNA probes, we organize 256 shades of red, green, and blue intensities that can be superimposed and merged. In so doing, hybridization alone is able to produce a large palette of 16 million colors or 24-bit color depth. Digital images can be reproduced with high fidelity at the micrometer scale by using a simple process that assigns sequence to any RGB value. Largely automated, this approach can be seen as miniaturized DNA-based painting.

An 8-bit monochrome palette of fluorescent nucleic acid sequences for DNA-based painting.

The ability to regulate, maintain and reproduce fluorogenic properties is a fundamental prerequisite of modern molecular diagnostics, nanotechnology and bioimaging. The sequence-dependence of the fluorescence properties in fluorophores commonly used in nucleic acid labelling is here being exploited to assemble a color scale in 256 shades of green Cy3 fluorescence. Using photolithography, we synthesize microarrays of labeled nucleic acids that can accurately reproduce 8-bit monochrome graphics by mapping color to fluorescence intensity and sequence. This DNA-based painting approach paves the way for a full RGB scale array fabrication process.

Sequence-dependence of Cy3 and Cy5 dyes in 3' terminally-labeled single-stranded DNA.

Fluorescence is an ideal tool to see and manipulate nucleic acids, and engage in their rich and complex biophysical properties. Labeling is the preferred approach to track and quantify fluorescence with nucleic acids and cyanine dyes are emblematic in this context. The fluorescent properties of cyanine dyes are known to be sequence-dependent, with purines in the immediate vicinity increasing the fluorescence intensity of Cy3 and Cy5 dyes, and the ability of nucleobases to modulate the photophysical properties of common fluorophores may influence fluorescence measurements in critical assays such as FISH, qPCR or high-throughput sequencing. In this paper, we comprehensively map the sequence-dependence of Cy3 and Cy5 dyes in 3'-fluorescently labeled single-stranded DNA by preparing the complete permutation library of the 5 consecutive nucleotides immediately adjacent to the dye, or 1024 sequences. G-rich motifs dominate the high fluorescence range, while C-rich motifs lead to significant quenching, an observation consistent with 5'-labeled systems. We also uncover GCGC patterns in the extreme top range of fluorescence, a feature specific to 3'-Cy3 and Cy5 oligonucleotides. This study represents the final piece in linking nucleotide identity to fluorescence changes for Cy3, Cy5 and fluorescein in all 3', 5', single-stranded and double-stranded DNA formats.

In silico modelling of DNA nanostructures.

The rise of material science and nanotechnology created a demand for a next generation of materials and procedures that can transcend the shaping of simple geometrical nano-objects. As a legacy of the technological progress made in the Human Genome Project, DNA was identified as a possible candidate. The low production costs of custom-made DNA molecules and the possibilities concerning the structural manipulation triggered significant advances in the field of DNA nanotechnology in the last decade. To facilitate the development of new DNA nanostructures and provide users an insight in less intuitive complexities and physical properties of the DNA folding, several in silico modelling tools were published. Here, we summarize the main characteristics of these specialised tools, describe the most common design principles and discuss tools and strategies used to predict the properties of DNA nanostructures.

Multiscale Visualization and Scale-Adaptive Modification of DNA Nanostructures.

We present an approach to represent DNA nanostructures in varying forms of semantic abstraction, describe ways to smoothly transition between them, and thus create a continuous multiscale visualization and interaction space for applications in DNA nanotechnology. This new way of observing, interacting with, and creating DNA nanostructures enables domain experts to approach their work in any of the semantic abstraction levels, supporting both low-level manipulations and high-level visualization and modifications. Our approach allows them to deal with the increasingly complex DNA objects that they are designing, to improve their features, and to add novel functions in a way that no existing single-scale approach offers today. For this purpose we collaborated with DNA nanotechnology experts to design a set of ten semantic scales. These scales take the DNA's chemical and structural behavior into account and depict it from atoms to the targeted architecture with increasing levels of abstraction. To create coherence between the discrete scales, we seamlessly transition between them in a well-defined manner. We use special encodings to allow experts to estimate the nanoscale object's stability. We also add scale-adaptive interactions that facilitate the intuitive modification of complex structures at multiple scales. We demonstrate the applicability of our approach on an experimental use case. Moreover, feedback from our collaborating domain experts confirmed an increased time efficiency and certainty for analysis and modification tasks on complex DNA structures. Our method thus offers exciting new opportunities with promising applications in medicine and biotechnology.