Séralini, roi de la com’ (2)

Les critiques de l’étude de Séralini sur les rats nourris au maïs RR et abreuvés au RoundUp continuent et proviennent du monde entier.

Boing Boing, déplore la campagne de relation publique mise en place par l’équipe de Séralini:

What’s not common: Forcing journalists to sign non-disclosure agreements promising to not show the study they’re reporting on to any independent researchers or outside experts. If you’re trying to make sure your publication runs a story on the study right when the embargo lifts, but you can’t show the study to any third-party experts before the embargo lifts, then the story you run is going to (inevitably) contain only information the authors of the study want you to talk about. It ceases being journalism and becomes PR.

Idem pour Carl Zimmer, plus connu pour ses billets sur l’évolution.

Séralini se fait à nouveau démonter dans Forbes. On y lit que Séralini et al. sont soutenus par The Sustainable Food Trust qui a mis en ligne un pdf répondant aux critiques sur la science de son Scientifique international de l’année 2011:

We might also look at the people pushing the paper. The Sustainable Food Trust. To a reasonable degree of accuracy this seems to be the militant wing of the Soil Association. For those of you who don’t know your British hippies this is essentially the British trade union for organic farmers. Yes, with all the nonsense about homeopathic treatment for animals although most of them do stop short of having to bury a cow horn in the dung pile by moonlight (no, really, there is a wing of the movement which insists that this helps in some manner).

[...]

If we are to be more serious we would point out that truth in science is not shown by peer review. Rather, by replication. Whether of the experiment itself or of the manipulation of the data as it is presented.

Le pdf de The Sustainable Food Trust fait beaucoup rire les intervenants du forum de Biofortified, dont pdiff qui est statisticien:

OMG! Call an ambulance! Seriously, I can’t breath. The stupid is too strong with these people. It is quite obvious they have not read or understood this study at all. Table 1 is simply a comparison of methods. It has NO data, let alone statistical comparisons. Grade this one Complete FAIL. [...] Quite frankly, if your expert statistician was closely involved with all this, then they suck at their job. Period. (Christ, hope I don’t get sued for that!)

If you are not talking about "a regular poison effect", then why the hell did they include a dose response series in the treatment design!!? We don’t know why the responses happened the way they did and speculating about hormonal disturbance is unsubstantiated. It does not matter what is expected in the completely unrelated issue of drinking a herbicide. we are talking about a GMO induced protein.

L’auteur de Forbes a également publié un billet soulevant une autre interrogation: quid de la présence d’OGM dans la ration des rats de labo aux É.-U.?

So to recap: the paper says that GM corn in lab chow for rats causes a huge rise in tumour growth. We have been feeding all lab rats in the US with GM containing lab chow for a decade now. We have not seen a huge rise in tumour growth in rats so fed.

Sur Twitter, The Genetic Tomato, légende correctement les graphes et photos de Séralini:

http://twitter.com/Gentomaat/status/249022917273346048/photo/1

Monsanto a répondu aux conclusions de l’étude et ça fait mal:

This study does not meet minimum acceptable standards for this type of scientific research, the findings are not supported by the data presented, and the conclusions are not relevant for the purpose of safety assessment.

Du côté français l’étude du CRIIGEN se fait écorner par le professeur Narbonne, toxicologue de métier pourtant très critique envers les OGM.

Au final, l’étude de Séralini ne lui apportera pas le prix Nobel mais servira d’exemple à ne pas suivre pour les étudiants comme le rappelle le blog d’un labo de statistiques.

I am grateful for the authors for publishing this paper, as it provides a fine case study for teaching a statistics class about poor design, analysis and reporting. I shall start using it immediately.

 

 

 

About these ads

Laisser un commentaire

Entrez vos coordonnées ci-dessous ou cliquez sur une icône pour vous connecter:

Logo WordPress.com

Vous commentez à l'aide de votre compte WordPress.com. Déconnexion / Changer )

Image Twitter

Vous commentez à l'aide de votre compte Twitter. Déconnexion / Changer )

Photo Facebook

Vous commentez à l'aide de votre compte Facebook. Déconnexion / Changer )

Photo Google+

Vous commentez à l'aide de votre compte Google+. Déconnexion / Changer )

Connexion à %s