In conjunction with the development of a new microarray platform (product name: matriXarray) we are currently developing an integrated workflow concept that will significantly decrease the time required for microarray design, delivery, processing and interpretation of results. The matriXarray platform includes an automated hybridizer/reader instrument (parallel processing of up to 6 chips), and innovative software tools for target selection, custom chip design, probe design, image analysis and data interpretation.
Array technology:
The matriXarray technology allows the in situ synthesis of oligonucleotides on a solid support by electrochemical placement of nucleotide monomers at specific locations (electrodes) of a substrate. Electrodes can be individually addressed and a virtual flask technology, allowing amidite de-protection only above the electrodes, confines the coupling of the next nucleotide to the exact location on the chip. With this technology the chip design is software controlled and can be altered for each individual chip. The density of the chip is scalable from 1170 sites/cm2 to 976800 sites/cm2. Due to classical oligonucleotide synthesis chemistry (acidic de-protection of DMT) synthesis efficiency is greater than 98% allowing the synthesis of up to 50mers.
The matriXarray platform is not limited to DNA applications. Protein analysis and protein/protein interactions are potential applications for the future.
Experimental Planning: Target selection for (custom) chip design:
Target selection is based on a linguistic software tool for gene selection. The tool is the combination of automatic analysis of the data sources in combination with the manually added value to the data by the expert. In respect to the enormous amount of data to be analyzed (e.g. Medline > 11x10e6 references) only automatic methods will allow for a comprehensive analysis. The manual step will only be performed on an extremely reduced and preprocessed data set.
e-Commerce custom chip ordering system:
The matriXarray e-Commerce portal allows users to order custom matriXarrays by sending lists of gene or probe sequences. Alternatively, customers can choose from a categorized gene shopping list (categories created with the target selection software) and thus configure a custom matriXarray, e.g. for an oncology array. Array delivery is approximately one week after placing the order. In addition to the array additional information for each gene (collected from various public and optionally proprietary databases) is provided with each matriXarray and feed into the data interpretation software.
Probe design software:
For custom arrays probes need to be chosen and synthesized on the matriXarray. The matriXarray probe design software is used to design optimized capturing probes for use with the platform. The designed probes correspond to the following requirements: -minimize possible secondary structures (minimum self-complementarity), -minimize cross-hybridization with non-target RNA or DNA and other sequence data (minimum cross-complementarity).
Image analysis software:
An easy-to-use image analysis software is provided with the matriXarray platform. The software is designed around a relational database system with a client-server architecture. Thus several hybridizer/reader instruments that process and readout matriXarrays can be connected to the database. Image analysis is performed with an graphical interface that includes flagging, artifact detection and correction, and normalization to spike-in controls. Grid-alignment is automated, facilitated from the fact that the matriXarray technology is based on silicon wafers with fixed feature size and distance. All results of the image analysis are stored in the matriXarray relational database.
Data interpretation software:
The matriXarray software for data analysis and interpretation includes advanced annotation management to add information to genes, gene lists and experiments. Annotations can be user defined in addition to the annotations provided with the matriXarray layout file (e.g. abstracts, protein structure, functional category available for each gene). This concept of bundling information in a regulated way (i.e. with a defined vocabulary), in combination with wizards that guide the user towards his scientific question, allows interpretation of microarray experiments in an efficient way. Established clustering algorithms (hierarchical trees, self-organizing maps) are combined with advanced visualizations (e.g. functional projection) that combine clustering results with gene and experiment annotation information. All information from matriXarray experiments (including experiment protocols, user information, date, chip-id, sample and probe information, etc.) is stored in one relational database with advanced user-management. Identified lists of genes can be used for a new matriXarray custom design.
