Figure 2

Flowchart of the data analysis. Five different cDNA libraries were sequenced and the reads were used to generate three different data sets. First, the data set “Highly expressed transcripts” were retrieved by assembling the reads of each library into isotigs (ranscripts) and normalizing the read counts. Using this approach, the 500 most highly expressed transcripts in each library were retrieved (the “Top 500” data set). Second, the data set “Differentially expressed UniRef50 clusters” was obtained by matching the isotig sequences using BLASTX [25] to UniRef50 clusters [26]. The procedure organized the isotigs into putative orthologs for which expression levels could directly be compared between the five libraries. Third, to identify the data set “Host-specific gene expression” the reads from the two libraries of A. oligospora were mapped to the genome sequence of this fungus [19]. Ao(Mh) denotes A. oligospora and M. hapla; Ao(Hs), A. oligospora and H. schachtii; Ad(Mh), A. dactyloides and M. hapla; Ad(Hs), A. dactyloides and H. schachtii; and Mc(Hs), M. cionopagum and H. schachtii. Further details of the libraries are shown in Table 2.