{"id":20489,"date":"2022-09-01T18:29:51","date_gmt":"2022-09-01T16:29:51","guid":{"rendered":"https:\/\/www.wjst.de\/blog\/?p=20489"},"modified":"2025-02-25T12:50:13","modified_gmt":"2025-02-25T10:50:13","slug":"the-amyloid-western-blot-schrag-vs-lesne","status":"publish","type":"post","link":"https:\/\/www.wjst.de\/blog\/sciencesurf\/2022\/09\/the-amyloid-western-blot-schrag-vs-lesne\/","title":{"rendered":"The amyloid Western blot: Schrag vs Lesn\u00e9"},"content":{"rendered":"

The amyloid analysis published in Nature<\/a> has been commented at PubPeer<\/a> and also earned a commentary\u00a0 of Charles Piller in\u00a0 Science. His “Blots on a field<\/a>” news story is leading now even to an expression of concern by Nature.<\/p>\n

\n

The editors of Nature have been alerted to concerns regarding some of the figures in this paper. Nature is investigating these concerns, and a further editorial response will follow as soon as possible.<\/p><\/blockquote>\n<\/div>\n

IMHO there are many artifacts including horizontal lines in Fig 2 when converting the image to false color display. I can not attribute the lines to any splice mark\u00a0 and sorry – this is a 16 year old gel image.<\/p>\n

Basically an eternity has passed in terms of my own\u00a0 camera history with 5 generations from the Nikon D2x to the Z9.<\/p>\n

So don’t expect any final conclusion here as long as we cannot get the original images.<\/p>\n

\"\"<\/a>
false color display of Fig 2 a “the presence of 8 M urea did not alter the electrophoretic pattern of A\u00df oligomers in extracellular-enriched extracts from 12- to 20-month-oldbrains ofTg2576\u00fe\/2mice that were probed with 6E10 antibodies.”<\/figcaption><\/figure>\n

<\/p>\n

Similar bands are worrying of course, but as Ashe writes on PP “bands that migrate close to each other may differ in intensity but appear similar in shape”.\u00a0 Unfortunately the Piller article did not respond to this argument raising doubts not only on the Lesn\u00e9 paper but also on the Schrag analysis in Science.<\/p>\n

 <\/p>\n

\"\"<\/a>
Fabrication analysis by as reported in https:\/\/www.science.org\/content\/article\/potential-fabrication-research-images-threatens-key-theory-alzheimers-disease I have no idea why they left the black vertical bar there?<\/figcaption><\/figure>\n

 <\/p>\n

According to Piller Schrag had only 4 weeks of PP experience with image analysis and is using here a method of Western blot alignment that has never been validated before – most likely a manual analysis of undocumented pixel shuffling.<\/p>\n

As Schrag did not respond to my email for technical details, I am trying now to repeat his analysis. So we read the images first and find the contours of the bands.<\/p>\n

\"\"<\/a>
Analysis of the 3 and 4mer band of Fig 2 A at 600px.<\/figcaption><\/figure>\n
\r\nim = cv2.imread( "schrag.jpg")\r\n# I am not adjusting background to keep the image as natural as possible\r\n# im = cv2.pyrMeanShiftFiltering(im, 25, 70)\r\n# converting to grayscale and applying threshold\r\nim = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)\r\nthresh = cv2.adaptiveThreshold(im,255,cv2.ADAPTIVE_THRESH_MEAN_C, cv2.THRESH_BINARY,101,10)\r\n# find contours\r\ncontours, _ = cv2.findContours(thresh, cv2.RETR_LIST, cv2.CHAIN_APPROX_NONE)\r\nfor i in range(len(contours)):\r\n    x,y,w,h = cv2.boundingRect(contours[i])\r\n    out = im[y:y+h,x:x+w]\r\n<\/pre>\n
\"\"<\/a>
Analysis of the 3 and 4mer band of Fig 2 A: Contours are similar but not identical.<\/figcaption><\/figure>\n
We then focus on the most distinct two bands on the right,\u00a0 correct their size and adjust contrast as done by Schrag.<\/p>\n
\r\n# change crop to make out1 comparable to out2\r\n# compressing eg cv2.resize(i2, (h,w), interpolation = cv2.INTER_AREA) seems to invasive\r\nout1 = out1[0:32,2:175]\r\nh,w = out1.shape\r\n\r\n# Michelson contrast\r\ndef getcontrast(im):\r\n    min,max =(int(np.min(im)),int(np.max(im)))\r\n    return( (max-min)\/(max+min) )\r\nprint( getcontrast( out1) ) # 0.31\r\nout2 = cv2.addWeighted(out2, 1.2, out2, 0, -50)\r\nprint( getcontrast( out2) ) # 0.33 which is acceptable\r\n\r\n# display both bands\r\nim = np.ones([w,h,3], dtype=np.uint8)\r\nim[:,:,0] = 255-out1\r\nplt.imsave("out1.jpg", im)\r\nim = np.ones([h,w,3], dtype=np.uint8)\r\nim[:,:,1] = 255-out2\/0.5\r\nplt.imsave("out2.jpg", im)\r\n# heatmap seems overkill to me but Spearman's R is nice to know\r\nr,_ = stats.spearmanr( im[:,:,0].flatten(), im[:,:,1].flatten() )\r\nprint(r) # 0.92\r\n# no change also after combining neighboring pixel\r\nim = cv2.resize(im, (int(w\/4), int(h\/4)), interpolation=cv2.INTER_NEAREST)\r\nstats.spearmanr( im[:,:,0].flatten(), im[:,:,1].flatten() )\r\n<\/pre>\n
\"\"<\/a><\/p>\n
\"\"<\/a><\/p>\n
These are the colorized bands basically as doen by Schrag.\u00a0 There is no need to construct the\u00a0 red + green = yellow overlay as the bands are clearly different. Anyway here is it just for completeness.<\/p>\n
\"\"<\/a><\/p>\n
I can’t replicate the results – neither the high correlation coefficient nor the shape of bands . There are also no splice marks although there should be some if Lesn\u00e9 would have used version Photoshop CS2 (9.0)<\/a> at that time.<\/p>\n
Is my alignment wrong? In don’t think so by visual eyeballing. Also comparing a larger area doesn’t change so much.<\/p>\n
\"\"<\/a> \"\"<\/a><\/p>\n
Keypoints also do not match, excluding largely any scaling and alignment issue.<\/p>\n
\"\"<\/a>
Side by side comparison of both bands after highlight clipping and spreading\u00a0 brightness values before constructing and matching keypoints. There are 27 in the first and 23 keypoints in the second band\u00a0 – so there is clearly a high similarity.<\/figcaption><\/figure>\n
\"\"<\/a>
Here is how a cloned band would look like.<\/figcaption><\/figure>\n
Of course I have also doubts on many Lesn\u00e9\u00a0 images including the duplication identified by Elisabeth Bik shown under #15<\/a>.<\/p>\n
Verdict: It seems that Piller has again produced another one of his dubious stories<\/a> while the original Lesn\u00e9 images need to be examined in more detail.<\/p>\n
1\/6\/2024<\/span><\/p>\n
I commented also at PubPeer<\/a> just before the paper was retracted<\/a> and asked for more respect.<\/p>\n
25\/2\/2025<\/span><\/p>\n
Piller has now even published a book<\/a> “Doctored
\nFraud, Arrogance, and Tragedy in the Quest to Cure Alzheimer’s”. As expected the book is not very well received, just to mention Lola Butcher at Undark<\/a><\/p>\n
His specific focus is on images \u2014 particularly Western blot images and micrographs \u2014 that bolster research findings. In \u201cDoctored,\u201d he takes a deep dive into two cases. The first is the story of Cassava Sciences …. The book also zeroes in on Alzheimer\u2019s researcher Karen Ashe, a University of Minnesota professor, and her colleague, neuroscientist Sylvain Lesn\u00e9 …<\/p>\n
Despite the worrisome picture they paint, the people who identified the problem images \u2014 and other experts Piller recruited to review their work \u2014 do not accuse their fellow scientists of fraud. Image-doctoring does suggest probable or apparent misconduct, Piller writes, but a definitive conclusion requires comparing the images with lab notebooks and other sources of data that were not available to the reviewers.<\/p>\n
Unfortunately, Piller\u2019s many tangents and the confusing organization of his material make for rough reading. For example, the first five chapters are devoted to detailed biographies of individuals who might be considered main characters of the book. As the narrative unfolds, though, the hundreds of minute details in those 50 pages are never shown to be particularly relevant.<\/p><\/blockquote>\n
or Ian Kremer<\/a>,\u00a0 Dave Weisman<\/a> or Dennis Selkoe<\/a><\/p>\n
Charles Piller\u2019s book contains errors of fact, distortions, and misleading messages to our patients…
\nFirst, Piller emphasizes some cases of fraudulent or badly executed research. Such work is grossly unacceptable … Second, there exists no \u201camyloid cabal.\u201d Many other features of Alzheimer\u2019s have been researched over decades … Piller\u2019s claim that an evidence-based focus on amyloid prevented other research is wrong. As of September 2024, the National Institute on Aging\u00a0had 68 active pharmacological trials on Alzheimer\u2019s; only 14 focused on amyloid.
\nThird, many investigators have chosen to work on amyloid because 13 specific lines of published evidence\u00a0support an early role for amyloid as a driver of tau and inflammation….<\/p><\/blockquote>\n
<\/div>\n
Science integrity should neither used as weapon<\/a> should not be exploited for financial gain, as happened in this instance.<\/p>\n\n
 <\/p>\n
\n  CC-BY-NC Science Surf , accessed 22.04.2026<\/span>\n <\/div>","protected":false},"excerpt":{"rendered":"
The amyloid analysis published in Nature has been commented at PubPeer and also earned a commentary\u00a0 of Charles Piller in\u00a0 Science. His “Blots on a field” news story is leading now even to an expression of concern by Nature. The editors of Nature have been alerted to concerns regarding some of the figures in this … Continue reading The amyloid Western blot: Schrag vs Lesn\u00e9<\/span> →<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[9],"tags":[3889,3887,3888,3891,3892,3890,218],"class_list":["post-20489","post","type-post","status-publish","format-standard","hentry","category-computer-software","tag-alzheimer","tag-bik","tag-christopher","tag-lesne","tag-piller","tag-schrag","tag-science"],"_links":{"self":[{"href":"https:\/\/www.wjst.de\/blog\/wp-json\/wp\/v2\/posts\/20489","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.wjst.de\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.wjst.de\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.wjst.de\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.wjst.de\/blog\/wp-json\/wp\/v2\/comments?post=20489"}],"version-history":[{"count":44,"href":"https:\/\/www.wjst.de\/blog\/wp-json\/wp\/v2\/posts\/20489\/revisions"}],"predecessor-version":[{"id":24700,"href":"https:\/\/www.wjst.de\/blog\/wp-json\/wp\/v2\/posts\/20489\/revisions\/24700"}],"wp:attachment":[{"href":"https:\/\/www.wjst.de\/blog\/wp-json\/wp\/v2\/media?parent=20489"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.wjst.de\/blog\/wp-json\/wp\/v2\/categories?post=20489"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.wjst.de\/blog\/wp-json\/wp\/v2\/tags?post=20489"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}