Analysis of a normalised equine cDNA library using deep sequencing. Equus caballus

While a first draft of the equine genome is available and predictions are made regarding resulting genes and proteins, little is known about the actual transcriptome. So far, published expressed sequence tags (ESTs) from different horse tissues were generally rather short (≤600bp) and hardly annotated, reflecting the problem that good cDNA libraries are very difficult to analyse. In this approach, we aimed to establish and analyse a normalised immune cell cDNA library (using freshly isolated and activated lymphocytes, NK cells, monocytes and DC). In particular, we wanted to test next generation sequencing combined with a series of bioinformatic approaches. The resulting cDNA library contained 2x107 clones of which 1056 were used for an initial Sanger sequencing and 4x106 for the deep sequencing analysis. Through the latter we obtained >29k sequences for which more than 5000 matches where found on the equine reference sequences. Additionally we could identify more than 3500 sequences which had matches on both - non-equine RNA sequences as well as the equine genome. In these we find both extensions of existing RefSeq models and novel mRNAs alike. Less than 2% of sequences did not have any match in the mentioned databases. Overall design: 1 pooled set of samples from one animal analysed