Recently, the Journal of Alzheimer’s Disease (JAD) solicited their readership to vote on the most influential research articles in the field of Alzheimer’s disease (AD) in the last 5 years. The response to this exercise was overwhelming: approximately 300 articles were voted on to determine which articles were deemed the “top papers”. Readers voted on the articles and 50 articles were selected (http://www.j-alz.com/top50).

This exercise inspired us to write a comprehensive review of the field of recent AD research [1–3], a so-called “greatest hits” of AD research. We used the 300 nominated papers as a basis for our review series. We believe that this method of reviewing the literature is somewhat novel. Traditionally, academic reviews consist of a small number of authors deciding what articles should be included in the review, the review typically being of very narrow breadth, but considerable depth [4,5]. This traditional approach has a number of advantages and disadvantages. Notably, it makes logical sense for an “expert” to be the arbiter of the quality of certain research articles to include, especially in a field as controversial as AD. Secondly, the traditional approach allows for the reader to rapidly get up to speed on a given topic within the vast field of AD research. However, the traditional review approach suffers from a couple of notable drawbacks. Firstly, it is subject to the biases of the authors; for example, an author who has spent her career advancing the amyloid hypothesis may be loath to give sufficient credence to alternative hypotheses. Secondly, such a deep and narrow focus may result in a review resulting in “not seeing the forest for the trees”, so to speak. As we discuss in our review series, there is simply no longer any evidence that AD is caused by one specific factor. Rather, AD is caused by a myriad of factors all interacting with one another, their interplay leading to the symptoms and pathological hallmarks that we refer to as AD. In our opinion, a lot of effort has been wasted in the past arguing the amyloid versus tau hypothesis as opposed to investigating the complex interplay between these proteins and other factors such as inflammatory responses, as is now routinely done. We attribute this phenomenon to the silos formed in each research field. It is very easy to attend an international Alzheimer’s disease conference and interact with only researchers investigating, for example, AD pathology, with the topic of, say, AD imaging, being as remote as an entirely different pathology.

By contrast, our approach seeks to mitigate the limitations of traditional reviews. The selection of articles, and even topics, for our review was largely independent of our own opinions about the most promising AD research. This “crowd-sourced” approach relied on the broad expertise available in the AD research community to select topics and articles for review. Of course, we did have to select articles that provide background information in order to introduce more advanced articles. Our approach largely removed the biases that we hold from the selection process. (We are molecular biophysicists with expertise in amyloid-β.) Additionally, our approach allowed for a broad overview of contemporary AD research topics. In our review series, we noted a few topics that the readership seems to find especially promising (i.e., “hot”): amyloid-tau synergies, the prion hypothesis of AD, the effect of AD on the brain connectome, and advanced brain imaging. Additionally, contemporary diagnostic guidelines and the results of the latest clinical trials are always of interest to the clinician. These active areas of interest are of importance to both new and senior investigators as they allow investigators to tailor their research to the contemporary unsolved questions in the field of AD research.

In addition to the advantages of our approach discussed above, our approach does suffer some limitations. Because of the breadth and the intent of our methodology, our reviews do not provide the depth that some other more contemporary reviews of a specific sub-field of AD may provide [6,7]. Additionally, we relied on the community of AD researchers to vote on papers that were mentioned. Sometimes crowds are wrong [8,9]. There is a possibility that some articles were nominated and included because they were “sexy” but lacked the scientific rigour that an experienced researcher could ferret out.

Our approach is not a replacement to traditional academic review authorship, but does demonstrate the power of collaboration when large groups of knowledgeable scientists converge allowing us to see both the “forest” and the “trees”.

Given this novel approach, it raises the question, are there any other alternative methods of review of scientific literature that may prove to be effective in the transmission of knowledge from author to reader?

Some time ago, I had idea of harnessing artificial intelligence (AI) as a method of producing individualized academic reviews. (If you’re into AI feel free to contact me about collaboration.) This “artificial” type of review may not be published, but instead, each reader could determine for themselves what topics they wanted covered in the review. An interface would work something like this: a user would type “Alzheimer’s disease” into a field in the interface. The interface would return a list of popular topics within AD that could be checked by the user. The user may be interested in, let’s say, inflammation, amyloid, and treatment, and select those topics on the interface. The program would than search the literature and select the most influential papers discussing those topics. (A weighting algorithm would weight papers to select by citations and recency.) Now here would be the difficult part: the program would have to mine the papers selected for the relevant data contained therein and summarize and organize the raw data into cohesive sentences and paragraphs. Voila! An instant, personalized, review covering only the topics you want to know. In theory, this could render the academic review obsolete.

In the meantime, reviews will continue to provide investigators with the most pertinent information at their fingertips.

References
[1] Hane FT, Lee BY, Leonenko Z (2017) Recent progress in Alzheimer’s disease research, part 1 : pathology. J Alzheimers Dis 57, 1–28.
[2] Robinson M, Lee BY, Hane FT (2017) Recent progess in Alzheimer’s disease research, part 2: genetics and epidemiology. J Alzheimers Dis 57, 317–330.
[3] Hane FT, Robinson M, Lee BY, Bai O, Leonenko Z, Albert MS (2017) Recent progress in Alzheimer’s disease research, part 3: diagnosis and treatment. J Alzheimers Dis 57, 645-665.
[4] Drolle E, Hane F, Lee B, Leonenko Z (2014) Atomic force microscopy to study molecular mechanisms of amyloid fibril formation and toxicity in Alzheimer’s disease. Drug Metab Rev 46, 207–223.
[5] Hane F, Leonenko Z (2014) Effect of metals on kinetic pathways of amyloid-β aggregation. Biomolecules 4, 101–116.
[6] Heppner FL, Ransohoff RM, Becher B (2015) Immune attack: the role of inflammation in Alzheimer disease. Nat Rev Neurosci 16, 358–372.
[7] Nasica-labouze J, Nguyen PH, Sterpone F, Berthoumieu O, Buchete N, Simone A De, Doig AJ, Faller P, Garcia A, Laio A, Li MS, Melchionna S, Mousseau N, Mu Y, Paravastu A, Pasquali S, Rosenman DJ, Strodel B, Tarus B, Viles JH, Zhang T, Wang C, Derreumaux P (2015) Amyloid β protein and Alzheimer’s disease : when computer simulations complement experimental studies. Chem Rev 115, 3518–3563.
[8] Flegenheimer M, Barbaro M (2016) Donald Trump is elected president. New York Times A1.
[9] Wolfe-Simon F, Switzer Blum J, Kulp TR, Gordon GW, Hoeft SE, Pett-Ridge J, Stolz JF, Webb SM, Weber PK, Davies PCW, Anbar AD, Oremland RS, Hille R, Lane TW, Morel FM, Wolfe-Simon F, Davies PCW, Anbar AD, Rosen BP, Baer CD, Edwards JO, Rieger PH, Oremland RS, Stolz JF, Hollibaugh JT, Blum JS, Bindi AB, Buzzelli J, Stolz JF, Oremland RS, Takeuchi M, Makino W, Cotner J, Sterner R, Elser J, Mandelstam J, Smith PG, Pickering IJ, Holbrook S, Dickerson R, Kim SH, Oremland RS, Stolz JF, Quillaguaman J, Delgado O, Mattiasson B, Hatti-Kaul R (2011) A bacterium that can grow by using arsenic instead of phosphorus. Science 332, 1163–1166.