Correction to Shavit Grievink L, Penny D, Hendy MD, Holland BR: LineageSpecificSeqgen: generating sequence data with lineage-specific variation in the proportion of variable sites. BMC Evol Biol 2008, 8(1):317.
Since publication of our article , we discovered an error in the second example. For this example, we state in the paper, we used the program MrBayes  with the JC+I+Cov model. However, we now found that, albeit appearances, this model is not implemented in MrBayes . In fact, no combination of I+Cov (e.g. HKY+I+Cov, GTR+G+I+Cov) is currently implemented in MrBayes . Instead, the program ignores the I parameter, so tree reconstruction in this example was therefore effectively done using the JC+Cov model. This does not affect the conclusion of our paper that phylogenetic estimation can be misleading for sequence data simulated with lineage-specific properties.