1. You covered this one. Cherry-picking data and an irrational exclusion of humans from all other vertebrates.

2. Horribly painful captioning, although that was probably not your first thought! They should have put the species names in parentheses or something more readable. The species are listed as a "phylogenetic sequence", which sounds to me like a linear route of descent. Hmm. The "phylogenetic sequence" should be just as accurate if we swap the yeast to the far right, since fungi and animals are sister groups. The species are ordered in a way that's supposed to show an evolutionary progression from lower to higher, which, of course, is an imaginary distinction. Since only one species is shown for each I wouldn't be surprised if more cherry-picking is going on.

3. Usually when you have a line graph that means there is some sort of logical progression from one point on the X-axis to the next. You can show percent completion of a reaction at different temperatures with a line graph, but percent alkylation at each site in a DNA strand should be shown with bars. This is a collection of individual species. This should be a bar graph.

4. Ow. Need I say more? Apparently the lepidosaurs diverged from the amniotes before the archosaurs did, and archosaurs and synapsids are sister groups. That's just the beginning of the problems with this chart. For instance, I see now that evolution has culminated with the human species, the evolutionary process can cease.