Over the past few months, there has been a great deal of criticism of the work of stem cell researcher Jacob Hanna, especially on PubPeer.
Now, Hanna has responded in the comments of the PubPeer entries for a number of papers. He has also posted a number of PDFs, including his PhD thesis, and correspondence with scientists who have been critical of his work.
He also wrote a response to a criticism published on bioRxIV, a non-peer reviewed repository of biology pre-prints.
One of Hanna’s comments on PubPeer is a summary of the issues with his Blood paper on PubPeer, blaming the figure errors on “medical trainees” who did the word while he was away at Mount Sinai. Here’s his explanation:
As we wrote, this will be corrected and explained in detail with the other issues very soon.
In brief, Figures 4,5,6 and 7 in the blood Manuscript were conducted and assembled by medical trainees (M.D., S.H. and N.W.) based on my instructions while I was away for 6 months at Mount Sinai NYC (end of 2004) for a medical student exchange program. Unfortunately, this led to mistakes in data presentation and use of wrong files, which i still take full responsibility for as their supervisor and first author of the paper. This also relates to mistakes in my 2004 JCI and JI papers, revised during the same period (Corrections for the latter have been submitted and accepted).
Fortunately, as the TAP2 deficient patients are still healthy due to the dedicated care of their physician, and as we have all ethical approvals still in place, we have repeated these experiments again in order to resolve this. While we have more work to do, all the results we have obtained so far maintain the validity of our claims and conclusions (as expected from a stable genetic mutation carrying cells, but always comforting of course:)), and we will approach Blood journal with a request for correction soon once we conclude conducting our final tests.
I apologize for the mistakes, but my main goal is to correct the scientific record and re-check the validity of the reported findings, and it clearly remains unchanged based on the data we have obtained so far.
It seems to be becoming more common for high-profile scientists to jump into the fray of anonymous post-publication review, which we think is a good thing. We look forward to the official corrections.
If it continues like this, PubPeer should be awarded the Nobel Peace Prize. Seriously. PubPeer has already done more for better science and therefore a better world to live in than it would have ever been expected when the project was started.
How is Prof Sarkar’s subpoena doing, by the way?
The positive side (which we rarely see, especially if we consider the ICMJE clauses of authorship):
“which i still take full responsibility for as their supervisor”
The constant face-saving line given by far too many cases questioned at PubPeer:
“all the results we have obtained so far maintain the validity of our claims and conclusions”
Hopefully those temporarily abandoned medical trainees are now excellent physicians and will no longer be condemned to collating data for the assembly of tricky figures for scientific papers. Here is what I’ve gleaned from PubPeer commentators on the Blood article now being fixed up
https://pubpeer.com/publications/C278F3DE939616C4ADBDB9C15DB268
Fig. 2
Plots for cells sorted by protein expression in 2005, some of which appeared a tad precociously in 2003
http://i.imgur.com/ZMEiqRq.png
In the same figure: Healthy donor CANK/CCR2 = Healthy donor CANK/CCR5
http://i.imgur.com/EASA0xn.png
Fig. 4
Plots for cells sorted by protein expression with vanishingly low probabilities of being so related:
Left column: (Patient A equals Patient C) is a subset of Healthy donor
2nd Column: Healthy donor now equals patient B
Right Column: Healthy donor is now partially overlapping with Mother
http://i.imgur.com/MYDeR8S.png
Fig. 5
B7-H3 of 2004 vintage NK cells had, by 2005, turned into CCR2 of a patient with an inherited disease
http://i.imgur.com/zZeHu5n.png
Fig. 7
Plots for cells sorted by protein expression that are related, despite somw of them being severely chopped up
http://i.imgur.com/CWDQyx2.png
Did I get anything wrong? Did I miss anything?
I think a bit more explanation is required as to how the combination of “my instructions” and the use of the “wrong files” led to the inclusion of data that seem already to have appeared in two previous publications.
Fig. 2 isn’t mentioned in the comment above. Hopefully the correction will also address that.
Presumably
S. H. = Suhair Hanna
M. D. = Maher Deeb
N. W. = Noam Weizman
Have they admitted “using the wrong files”? Do they know about this discussion?
Finally, if the experiments need to be redone, that means that original data supporting the published conclusions do not exist. And there is therefore no proof that it ever existed.
It seems to me that a lot of scientists have been bitten by not having sufficient IT experience for what their fields now involve. Time was, you could paste your data into your lab notebook and label it diligently, and there would be essentially no chance that you or someone else would accidentally get a data set that wasn’t the right one. That quality of record keeping, applied to digital data, requires practices that are different from both the physical practices well-known from science class and the ad-hoc file handling that most people use in their everyday computer interaction. Too often, they end up with the digital equivalent of a scrap of paper left somewhere in the lab, hopefully near the right experiment, and mistakes usually don’t happen, but sometimes do.
It now appears that the once-celebrated “Mbd3” deterministic reprogramming only works in cells from the Hanna lab.
In 2013, the reports of Mbd3 inhibition causing 100% reprogramming and solving the problem of low reprogramming efficiency was thought to revolutionize the field. However, it is now 2015 and no one can reproduce this startling finding.
The mystery of how the figures were generated for that paper is finally being solved.
This paper is also part of Hanna’s Ph.D. thesis. Doctoral dissertations consist of a Ph.D. candidate’s original work. If a thesis includes data produced by someone else, that should be stated. Looking through the PDF, none of the people referenced in the comment above (M.D., S.H., N.W.) are credited. Either they should have been acknowledged or Hanna is responsible for that work.
To quote Lady Bracknell (thus saving Fernando Pessoa the trouble): “To duplicate one figure … may be regarded as a misfortune; to copy more, looks like carelessness.”
It is remarkable how often ‘mistakes’ and use of ‘wrong data’ seem to support the main findings of papers discussed on RW.
I note that the link to the response by Dr Hanna is not functional, at least for me.
@Daniel Barkalow has identified a major problem. The investment in IT infrastructure in universities generally has focussed on communication and information, but not on digital data, which has been left to PIs to sort out. The meta data that provide the link between digital data and experiment will often be non-existent, unless the PI really pushes people on this. Student training, from undergraduate on, also lags on this front. Open data may help solve the problem, since it will force archiving of data in readable formats.
@Mike Briggs, you have saved both Fernando and myself some words!
Regardless of justifiable concerns by the community, at least we have author engagement, which is a great step forward from deafening silence!
why the waste of time in excessive intellectual fruitless discussion on the meaning of Fig X and Y, if a person is caught with his pants down he blames the loose belt, the rest is speculation, excuses, misunderstanding, unintended oversight, and so on,. What is important is it reproducible or not, or did it it harm any patient.
I am afraid there is something I don’t understand. Dr Hanna says that he was away in the US at the end of 2004. According to his CV, previously available on his lab website but recently removed ( http://i.imgur.com/TO3OmPw.jpg – need to click to magnify), the duration of his fellowship from the Hebrew University to stay at the Mount Sinai Hospital corresponds to 2004, rather than 2004-2005. Compared to other fellowships that he held that spanned over two different years, it appears from his own CV that he only stayed in the US during 2004. The paper was submitted to Blood on the 1st March 2005, so it appears that Dr Hanna had two spare months to prepare and check his own work himself, work that was shortly to be submitted. After that, there were another four months during which the paper was being reviewed and sent back to the authors, as it was only accepted on the 8th July. After that there were proofs that would (should?) have been carefully checked by the authors, as the final publishing date was the 15th of November that year. Were the medical trainees also preparing the images and addressing revisions while Dr Hanna was in the lab in 2005?
From the author’s note, one of the journals already agreed to publish a correction. But some of the duplicated figures and data were published in different journals. How does an erratum work in this situation, do all the journals review the material and agree?
It appears from Pubpeer that all journals involved accepted the corrections. Great news.
Jacob Hanna has made several breakthroughs in the stem cell field over the last 8 years. Simple inspection of these recent publications indicates his seminal contributions that are being used by leading labs including Surani, Gribnau, Graf and others:
1) http://www.cell.com/cell/abstract/S0092-8674%2814%2901583-9
2) http://www.ncbi.nlm.nih.gov/pubmed/23533168
3) http://www.ncbi.nlm.nih.gov/pubmed/25640760
4) http://www.ncbi.nlm.nih.gov/pubmed/24336202
5) http://www.ncbi.nlm.nih.gov/pubmed/25569111
….
Jacob Hanna is currently, and by far, the most innovative scientist in the iPS cell field. For those who know him personally, he is one of the most modest, generous and non-assuming people that one can meet. His work is revolutionary, and he will prevail.
“Jacob Hanna is currently, and by far, the most innovative scientist in the iPS cell field. For those who know him personally, he is one of the most modest, generous and non-assuming people that one can meet. His work is revolutionary, and he will prevail.”
Two definitions available on the internet for a Sock Puppet
1: A fake personality, usually a ‘friend’ or ‘sister,’ created by a drama queen/king for the sake of defending him/herself against others in an online forum.
2: Someone who might be an actual person but behaves like the above, defending someone who really deserves no defense.
From PubPeer’s FAQ: Please do not post about the scientists themselves. May I quote some anonymous texts from PubPeer threads that are relevant to this most modest, generous and non-assuming person?
https://pubpeer.com/publications/24172903
“Jacob Hanna blazed a trail through a jungle of dogmatic scientists.“
“It is an amazing achievement by the Hanna and Surani groups.”
https://pubpeer.com/publications/24048479
“You are rehashing your own generated misleading rubbish. Congratulations on joining the Austin Smith hunt dog team, where you can faithfully conduct your service to blindly support your master’s current dogma. I am sure this secures the path to your independence.”
“Oh please, this is so childish and misleading. See wonderful response from Jacob Hanna released today”
https://pubpeer.com/publications/073BF7497C8B0D6FF50CE1569BC422
“Unfortunately this paper was scooped by Hanna’s neural stem cells that he called naive human stem cells (Hanna et al, PNAS 2010).
Not surprisingly, Hanna scooped others again (Gafni et al) with primed human stem cells that he called naive human stem cells.
How many different versions of naive cells can this PI make?”
https://pubpeer.com/publications/1319F6076BEA40C1BF6C22C96A4AEB
“This Hanna and Jaenisch paper started it all.”
Jacob Hanna is the kindest, bravest, warmest, most wonderful human being I’ve ever known in my life.
That is very true.
A new paper just published in Nature by Marius Wernig (Stanford), independently confirmed the validity and authenticity of reprogramming obtained in Mbd3flox/- system.
http://www.nature.com/nature/journal/vaop/ncurrent/pdf/nature14274.pdf
Selected quote:
“We then wondered whether a similar intermediate population arises in high-efficiency reprogramming systems (12, 13). Published expression analysis of two high-efficiency systems showed transient CD73 upregulation, suggesting the presence of a similar intermediate (Extended Data Fig. 8a, b). We then characterized one of these systems, the Mbd3fl/− secondary MEF system, in greater detail (12). After confirming reported reprogramming efficiencies, we analysed this system by mass cytometry (Extended Data Figs 8, 9, 10). By day 3, fibroblast marker repression was evident, and CD73 was upregulated within this population (Extended Data Figs 8e and 9b). Within the CD73high/MEFSK4low population, CD49d (Itga4) upregulation was not apparent, but we noticed the emergence of a separate integrin, CD104 (Itgb4). By day 4, the major CD73high branch clearly overlapped with the CD104high branch and persisted into day 5. SSEA1high and CD326high expression was present on day 4, but clear co-expression was not seen until day 5. By day 9, CD73 and CD104 expression was dramatically reduced while CD326 and SSEA1 expression remained high. These data demonstrate a transient CD73high/CD104high population arises after donor cell program repression and before ESC marker acquisition, even in a highly efficient reprogramming system. Similar to CD49d and CD73, CD104 is not highly expressed in ESCs (Fig. 1a). And similar to viral reprogramming, adenosine treatment abolished reprogramming in the Mbd3 reprogramming system, albeit only at late stages, whereas compounds affecting CD49d function had little effect (Extended Data Fig. 4f).”
Have a look at https://pubpeer.com/publications/24048479 where the interpretation of Wernig’s data with the Mbd3+/- cells are questioned
Retraction: April 1, 2015
Jacob Hanna, Tsufit Gonen-Gross, Jonathan Fitchett, Tony Rowe, Mark Daniels, Tal I. Arnon, Roi Gazit, Aviva Joseph, Karoline W. Schjetne, Alexander Steinle, Angel Porgador, Dror Mevorach, Debra Goldman-Wohl, Simcha Yagel, Michael J. LaBarre, Jane H. Buckner and Ofer Mandelboim
Novel APC-like properties of human NK cells directly regulate T cell activation
J Clin Invest. 2015;125(4):1763–1763. doi:10.1172/JCI81527.
Copyright © 2015, American Society for Clinical Investigation
….and republished in a corrected form on bioRxiv: http://biorxiv.org/content/early/2015/03/30/016816
Corrected 2004 JCI paper by Mandelboim group has been republished on bioRxiv: http://biorxiv.org/content/early/2015/03/30/016816
Correction published in Blood journal as indicated by Hanna.
http://www.bloodjournal.org/content/125/16/2583.2