There was a 91% probability that nurse-led follow-up was cost-effective compared to standard follow-up, discarding quality-of-life effects (willingness to pay €0, Figure 3).
Low quality reads were discarded (quality limit of 0.05).
We used Tophat (Kim et al., 2013; Trapnell et al., 2009) version 2.0.6 to align the reads, discarding low-quality alignments (quality score below 1).
After discarding low quality reads, 607.2 million high quality reads (~83.3% of total reads) were used for further analysis.
After discarding low quality sequences, more than 17 million clean reads were obtained for each tissue.
After discarding low quality reads and sequence counts lower than 3 in each library, clean reads were produced.
After discarding low quality sequences, each library retained 71 80% sequences with an average of 73% of the total 11,520 phage producing high quality sequence data.
After discarding low quality reads, adaptor sequence, contaminating mRNA, rRNA, tRNA and snoRNA sequences by Illumina Pipeline filter (Solexa 0.3), 12,744,815 clean reads (87%) were obtained.
The reads were then processed by discarding low quality and vector sequences using Lucy v1.19p [ 64] with default parameter values, by removing poly-A tails using EMBOSS trimest [ 65] and by discarding sequences with lengths under 100 bases.
Therefore, as a prelude to developing a reliable physical map of B. rapa, it is worth discarding low quality or problematic data before the fingerprint assembly to avoid chimeric contigs.
After discarding low quality sequences containing ambiguous nucleotides, sequences shorter than 18 nucleotides, and sequences counts below 3 in both libraries, 1,790,395 (WT) and 3,345,746 (MT) reliable sequences were remained for analysis.
