A major effort in 2008 was the re-arraying of RIKEN (FANTOM I to III) clones to provide a single chip representing more than 22,500 murine genes. To date, almost 100,000 RIKEN clones have been processed and spotted on a set of four chips (200 chips per batch). Hybridisation information from more than 500 hybridised samples was used to identify clones that reproducibly yielded good hybridisation signals in at least three independent experiments. This information was used to re-array a non-redundant set of 22,500 clones that were subsequently processed for printing.

The reduction in the number of slides used for hybridisation will not only reduce the cost per experiment but also the time for analysis by a factor of three, allowing the processing of more samples. Currently we are trying to establish hybridisation of labeled RNA instead of cDNA; this will also substantially shorten the handling time and decrease the cost per experiment.

Processing of microarray data has been completely automated by combining several packages from the BioConductor project. We can now provide users with fully annotated lists of differentially regulated genes.

In the past six months, together with scientists from the Busslinger and Jenuwein labs we have set up the infrastructure for Next Generation Sequencing. Sample preparation and in-process quality control have been established and are currently being optimised.

We expect a throughput of up to two runs per week yielding up to 2.7 Gigabases per run depending on sample quality. By the end of this year, deep sequencing applications like ChIP-seq and sequencing of small RNAs will be offered as a service to scientists of both institutes.