The random sequences were generated by a random base generator that takes into account the predilection given to certain bases in the region under comparison.
To ensure allocation concealment, randomization sequences were generated by a computerized random treatment generator hosted on a dedicated trial Web site (http://peritonealvac.com).
Sequences were generated by using ABI3730 and MegaBase 1000 automatic sequencers and assembled by Phred/Phrap/Consed software [42], [43], [44].
Standards ranging from 1 to 25,000 sequences were generated by directional ligation of oligonucleotide "words" of standard length and GC content and then amplified by PCR.
CAA or SL sequences were generated by annealing paired oligonucleotides and inserted between the IRES and the firefly luciferase coding sequence using restriction sites SacII and NcoI.
Both genome sequences were generated by the US Department of Energy Joint Genome Institute.
The consensus sequences were generated by the COBBLER (Consensus Biasing By Locally Embedding Residues) program [36].
Tychosporium acutostipes sequences were generated by Ronan Bakker as part of his undergraduate honor's thesis.
The Xist consensus sequences were generated by the AnnHyb program (http://www.bioinformatics.org/annhyb).org/annhyb
Consensus sequences were generated by contiguation functions of BioEdit, Version 7.0.9 [29] or Geneious Version 5.0.2 [30] and manually edited, where necessary, by reference to the chromatograms.
These 5' leader sequences were generated by PCR and subsequently cloned into the plasmid vector (see Methods S1 for details of 5' leader sequences).
