Author: Raphaël Lévy

References for my talk at Gold 2018

Update: the slides are available (here: gold2018-Levy2) and there is a recording of a few minutes here.

Slide 1:

Slide 2: Nanotech is bs Tweet.

Slide 3: Calling Bullshit.

Slide 4:  Dinosaur. (from here)

Slide 5: Electron microscopy of Hela cells after the ingestion of colloidal gold; C.G. Harford, A. Hamlin, and E. Parker; 1957

Slide 6: The entry and distribution of herpes virus and colloidal gold in Hela cells after contact in suspension; M. A. Epstein, K. Hummeler, and A. Berkaloff; 1963

Slide 7:

Slide 8: The spherical nucleic acid paradox; D. Mason, G. Carolan, M. Held, J. Comenge, S. Cowman, and R. Lévy; 2015

Slide 9: Excerpt from email (shared with permission).

Slide 10:  Evaluation of SmartFlare probe applicability for verification of RNAs in early equine conceptuses, equine dermal fibroblast cells and trophoblastic vesicles;  S. Budik, W. Tschulenk, S. Kummer, I. Walter, and C. Aurich; 2017

Slide 11: SmartFlares fail to reflect their target transcripts levels; M. Czarnek and J. Bereta; 2017

Slide 12: Calcium-Binding Proteins S100A8 and S100A9: Investigation of Their Immune Regulatory Effect in Myeloid Cells; J. Yang, J. Anholts, U. Kolbe, J.A. Stegehuis-Kamp, F.H.J. Claas and M. Eikmans

Slide 13: SmartFlare catalog.

Slide 14:


The conference dinner chatter way of (not) correcting the scientific record

One of the common responses of senior colleagues to my attempts to correct the scientific record goes somewhat like this:

You are giving X [leading figure in the field] too much credit anyway. We all know that there are problems with their papers. We discussed it at the latest conference with Y and Z. We just ignore this stuff and move along. Though of course X is my friend etc.

This approach is unfair, elitist and contributes to the degradation of the scientific record.

First, it is very fundamentally unfair to the many scientists who are not present at these dinner table chatters and who may believe that the accumulation of grants, prizes, and high profile papers somewhat correlate with good science. That group of scientists will include pretty much all young scientists as well as most scientists from less advantadged countries who cannot get so easily to these conferences where the truth about scientific achievements is discussed between drinks at the end of a play-acting day (for inquisitive questions at the end of talks are of course also not the right way to act).

Second, it devalues fundamentally the role of the scientific record. We are basically accepting that it does not matter whether what gets published is right or wrong.  Here, I’ll insert an anecdote. I reviewed, a couple of years ago, a high profile review authored by a senior colleague in the field of nanoparticles. One of the figures highlighted a paper which I knew to be fundamentally wrong. In my review, I pointed that fact. The senior colleague did not dispute that the paper was flawed but he opted for keeping the figure, not discussing the fact the paper was wrong. His (post-modern) argument was that the  “concept” was important.

Dinner chatter is fine. But please also share your criticisms, e.g. via PubPeer.

(and by the way, if you could comment on our preprint on Spherical Nucleic Acids, that would be much appreciated)


The great answer to people saying that #preprints are not peer-reviewed

That perfect title is courtesy of (see tweet below)

On Monday (25/06), we will publish a preprint about the spherical nucleic acid technology. Our paper was prompted by the publication in Nature Biomedical Engineering of “Abnormal scar identification with spherical-nucleic-acid technology” by Yeo et al.

The great answer is… review them! I issued a call to review our preprint before it comes out and I have now sent the article to a number of colleagues across the world. I am very much looking forward to their comments good or bad. The comments will be posted on PubPeer. If you have some time on your hands this Friday or over the weekend to look at the paper, drop me an email and I will also send you a preview copy.

Yeo et al corresponding authors were provided with a copy of our preprint two weeks ago but unfortunately they have not responded. I hope they will post comments on PubPeer. We are planning to subsequently submit a version (hopefully improved thanks to the comments) to Nature Biomedical Engineering. It is however sometime rather difficult to debate the scientific literature through the official channels of traditional journals so this route via preprint will accelerate this important discussion.


Happy holidays, the preprint is out there!

This is a guest post by Marie Held. Credits are given to Andrew Plested for the title/timing of this post (see here).

It had been a long time coming but the preprint of our manuscript Ex vivo live cell tracking in kidney organoids using light sheet fluorescence microscopy is now available on BioRxiv. Together with the preprint, the associated large body of imaging data has been made publicly available in the online Image Data Resource (IDR) repository. Through the power of OMERO you can have a play with the data or even download it and then re-analyse it.

The imaging has been conducted on the Zeiss Z.1 Lightsheet microscope at the Centre for Cell Imaging at the University of Liverpool, whose ever helpful staff enabled us generating lots and lots of image data and providing an efficient infrastructure for data storage as well as support for image and subsequent data analysis. Of course this is only half the story because the imaging would not have been possible without generating the samples first: A big thank you also goes to the students and academic staff in the Institute of Translational Medicine at the University of Liverpool.

In this study we have generated organoids from dissociated and re-aggregated mouse embryonic kidney tissue and imaged them with a light sheet fluorescence microscope. The microscope optically sections the samples, therefore preserving the three-dimensional context of the sample throughout imaging. We have found organotypic kidney structures in the organoids and evidence for the maturation of cells to the point of forming glomeruli, the basic filtration unit of the kidney. A functional assay showed that the developed tubules display secretory function.

Most importantly though, we have also performed live imaging of organoids made from genetically tagged fluorescing cells. The light sheet microscopy setup combines an illumination that is perpendicular to the detection. Therefore, full frame images can be recorded rapidly and only the section of the sample that is recorded is illuminated, thus vastly reducing photobleaching and phototoxic effects that limit long-term live fluorescence imaging in wide field and in particular confocal scanning fluorescence microscopy. We have then tracked the fluorescing cells with the help of an automated algorithm and subsequently analysed the generated tracking data. We have {started to} analyse the tracking data and can now quantitatively compare between experiments.

Yet, there is so much more that can be done with the images and data and we would love to see which ideas and approaches others might have, so please do not be shy to dig in and have a play. Be sure to let us know though.

Featured image caption: Organoid of mouse embryonic kidney cells formed following dissociation and re-aggregation of embryonic kidney rudiments. Yellow: Pax2+ cells indicating the metanephric mesenchyme, a prerequisite for nephron development, Red: Peanut-agglutinin staining basement membranes of the ureteric tree and developed nephrons.

Mind-boggling plagiarism of this blog

In January 2015, someone went to the effort of creating a fake raphazlab blog as the stripy nanoparticles controversy was descending from a scientific debate into the gutters of online discussions.

Fast forward three years. The Spherical Nucleic Acids controversy is slowly heating up. Chad Mirkin continues to win prizes after prizes, but he is unseemly asked to comment on the failing SmartFlare technology (the commercial name of the Spherical Nucleic Acids) by Dalmeet Singh writing for Chemistry World.

Dalmeet writes:

But Chad Mirkin, a chemist at Northwestern University in the US, who developed the precursor to SmartFlares, nanoflares, pointed Chemistry World to more than 30 papers, which, he says, have successfully used the technology.

Chemistry World contacted a number of groups that have used SmartFlares. Hirendranath Banerjee, a molecular biologist at Elizabeth City State University in North Carolina, describes SmartFlares as a ‘very novel and effective technique’, noting that it has been helpful in evaluating gene expression experiments in his lab.2

Now comes the mind-boggling part.

The introduction of Hirendranath’s paper (reference 2 above) is largely plagiarized… from this very blog. From the very first SmartFlare post on this blog, entitled How smart are the SmartFlares?

Below, is an excerpt from my post with, in red, the sentences that reference 2 copied.

RNA molecules play crucial roles in cells such as coding, decoding, regulation, and expression of genes, yet they are much more difficult to study. SmartFlares are nanoparticle-based probes for the detection and imaging of RNA in live cells. Could they become the GFP of the RNA world?

 Many certainly believe this to be the case. SmartFlare ranked second in TheScientist top ten 2013 innovations, with one of the judges, Kevin Lustig, commenting “These new RNA detection probes can be used to visualize RNA expression in live cells at the single-cell level.”  The following year, SmartFlare won an R&D100 award. The technology comes from Chad Mirkin’s lab at Northwestern University. Chad Mirkin is the winner of numerous prestigious prizes and a science adviser to the President of the United States. The scientific articles introducing the SmartFlare concept (under the name of Nano-Flare) were published in the Journal of the American Chemical Society in 2007, ACS Nano in 2009, etc. In 2013, the SmartFlare technology was licensed to EMD Millipore. Here is one of their promotional video:

For a molecular sensor to work, it needs a detection mechanism. The principle of the SmartFlare is explained from 0:45. A capture oligonucleotide (i.e. DNA) is bound to the gold nanoparticles. A reporter strand is bound to the capture strand. The reporter strand carries a fluorophore but that fluorophore does not emit light because it is too close to the gold (the fluorescence is “quenched”). In the presence of the target RNA, the reporter strand is replaced by the target RNA and therefore released, quenching stops, and fluorescence is detected.

This is plagiarism, with, in addition, a clear intent to deceive: whilst the article entire point appears to be the celebration of the SmartFlare technology, e.g concluding sentence (Thus Smart Flare novel gold nanoparticles could revolutionize the field of differential gene expression studies and drug discovery), the 2015 blog post was already doubting the validity of the technology.
I wrote to the Editor of the Journal who said that they would evaluate the claims and take some form of action. I contacted Hirendra Banerjee who declined to provide a statement; instead he issued a legal threat against the publication of this post.
Meanwhile, on Twitter
Conclusion of this exchange?

Drug Discovery 2017

This is a guest post by Marie Held reporting from the ELRIG conference held last week.

On 3rd-4th October I attended ELRIG’s flagship event, Drug Discovery 2017, in Liverpool. With around 250 participants, it was the largest of the ELRIG conferences yet. The spacious arrangement of the vendors and posters in the exhibition hall was a refreshing change. There was ample space to mingle, chat and discuss equipment on show.

On day one, I attended the Advances in Imaging stream (one of three parallel streams). The keynote lecture by Tony Ng covered a broad range of the spatial scale, stressing the importance of whole body imaging in cancer in combination with investigating the tumour microenvironment down to super resolution imaging of individual molecules. He outlined their attempts in predicting tumour metastasis enabled via immune system hijacking by the cancer cells. An important conclusion was that with the wealth of imaging methods and tracers being developed, we need standardisation and validation across facilities to bring them closer to the clinic, ultimately improving the lives of patients. The imaging methods discussed in the following six talks ranged from man to molecule, focussing on ever smaller features as the day went on. A transpiring theme was the generation of large amounts of data from different techniques and the associated challenge of deriving meaningful information. Machine learning and artificial intelligence were mentioned time and again as being part of that quest. The last scientific presentation, by Charlotte Dodson, focussed on twinkling enzymes, studying the conformational changes of kinases in disease and after treatment via single molecule spectroscopy. Throughout the imaging stream, twelve men contributed to the presentations, vendor snapshots and poster tasters and three women contributed to the stream. The other streams were a bit more gender balanced but only the workshop on Tuesday achieved a 50/50 split.

On the second day, I attended the Lab of the Future workshop presented by SiLA and ELRIG. The general consensus was that the lab of the future (whether you call it Lab 4.0, Industry 4.0 or something else) is an interconnected space in which smart machines are communicating with each other, running fully automated cycles of fabrication, screening and/or testing. Machinery that can be monitored if not controlled remotely via mobile device apps was mentioned multiple times. Smart products are uniquely identifiable, may be located at any time and “know” their own history, current status and alternative routes to achieving their target state. It left some of the audience wondering where innovation is going to come from. A lot of innovation is not based on a “Eureka” moment but rather lucky accidents or not quite sticking to the protocol and making mistakes. These instances are near on excluded in an automated lab. Another doubt that was raised was: Where is the space, if not need, for the scientist is in this fully automated lab? “He” has more time to think about the science and efficiency gains rather than processing the work. Unfortunately, the scientist was exclusively referred to as a “he” throughout the whole workshop, which irritated myself and another female member of the audience to the extent that it seemed appropriate to clarify that the scientist can be a female scientist. Unconscious discrimination is one of the reasons why there are still so few leading women in science. There was a conspicuous lack of women, both in the audience and in particular in the selection of session leaders, which were all male. It would be nice to see some female panel members in the future. Also, this year only one out of 12 session chairs throughout the whole conference were female.

Near on every panel member in the lab of the future workshop voiced that the interconnectivity should be down scalable to medium and small labs. As a member of the academic research community and a small lab, I felt somewhat left out though. We do not generally use automated machinery, never mind machinery connected to the internet of things. Often enough there is a piece of equipment, that has to be taken off the net entirely because the software is so outdated (and not supplier maintained anymore) that it has to run on an obsolete operating system posing a risk to the University network. That means we are in fact taking a step away from the lab of the future. The audience saw the responsibility with the industrial sector to come up with a solution and I am looking forward to seeing a change in the future. Also, electronic notebooks (find the same presentation here with audio comment) are already a standard in the industrial sector but the academic sector is severely lagging behind. Not all universities have specific guidelines on how to keep a paper lab book, never mind having a system of electronic lab books in place. The responsibility here lies in the academic sector to catch up but it might have to be a bottom up approach to induce a change.

The high point of the second day and probably the conference as a whole was the plenary keynote by Dr Nessa Carey asking whether we can fix big pharma. Her keynote was eloquent, inspiring and also entertaining. We can all do our bit to help fix big pharma. It is not the evil it is often made out to be. Millions of lives have been saved by pharmacological advances and still are being saved, however it does suffer from the worst PR there is.

Overall, I enjoyed the ELRIG Drug Discovery 2017 and am looking forward to the next instalments in London in 2018 and back in Liverpool in 2019.