The biggest was probably stating the results as facts in the result section and only as "more likely" in the conclusion. Sample size, taking only part of the ge

The biggest was probably stating the results as facts in the result section and only as "more likely" in the conclusion. Sample size, taking only part of the genomes. The code quality of their scripts on GitHub and especially the comments in them. Bad spelling and convulted sentences... there is a lot more have a look yourself.