ALMA Bioinformatica of Madrid has released its expression data analysis algorithm, the self-organizing tree algorithm, or SOTA, for commercial and public licensing. This algorithm “is a neural network that grows, adopting the topology of a binary tree,” the company said.
The algorithm, which the company said is suitable for large datasets, combines unsupervised hierarchical clustering and the supervised neural network. The clustering process is performed from top to bottom, and the algorithm includes a criterion to stop the growing of the tree based on the approximate distribution of probability using randomization of the original dataset.
The algorithm also allows users to obtain average gene expression patterns and enables them to map gene ontology data onto the hierarchical structure of clusters. The algorithm is available for Windows, Solaris, and Linux. A demo is available for downloads at: www. almabioinfo.com/products_software_ sota.html.
Ciphergen Biosystems has rolled out a new prototype protein chip with new surface chemistry. The surfaces are modified with homogenous polymeric coatings, the company said. When a laser is focused on the coatings, they develop ions without creating too much chemical noise, and the chip can be used to detect any binding to its surface when it is analyzed using surface enhanced laser desorption (SELDI). This new chemistry eliminates the need to add a matrix of solution as a separate step and enables detection of low-weight molecules, according to Ciphergen.