"Why not ... [submit] his own analysis of the public data ... explaining how it improves on existing analyses?" This is the meat of the matter. Scientists wri

"Why not ... [submit] his own analysis of the public data ... explaining how it improves on existing analyses?" This is the meat of the matter. Scientists write letters to journals all the time saying that they think an author overstretched. Those letters, in and of themselves, do not negate the previous research. Scientists also write original papers showing that a previous hypothesis is invalid. Those papers, if they are well written, are easy to publish, especially if they make the journal more interesting.