Libraries and research in a COVID-19 world - resource list and some thoughts
While I have been adjusting to the latest change in my personal life, the world has been grappling with the impact of COVID-19. Updated : 30th April 2020
Here are my list of interesting resources that caught my eye.
Tracking of library responses to COVID-19
Resources on working on a COVID-19 world
Opening up of resources by Publishers, Vendors and more
Datasets and analysis on COVID-19 articles
Does the COVID-19 crisis expose problems in the Scholarly Communication system?
Tracking of library responses to COVID-19
First off, if you are on Twitter, the hashtag #CovidLibrary appears to be the one to track, a lot of the resources here were found using that. (Harvested tweets)
Want to see how US academic libraries are responding to COVID-19?
Christine Wolff-Eisenberg, manager of surveys and research at Ithaka S+R, and Lisa Janicke Hinchliffe, professor and coordinator for information literacy services and instruction at the University of Illinois at Urbana-Champaign have you covered.
They first setup a Qualtrics survey on March 11 on how Academic Libraries in the US were responding to COVID-19 and as of writing they have over 700+ academic libraries in the US responding!
https://tinyurl.com/covidlibrary
You can see live data as responses come in, limit to ARL libraries and even limit to libraries that have closed and their strategies. Don't want raw data? The Ithaka blog provides summaries of the data so far.
From Canada? See here and here
From Australia? Australian GLAMs and COVID-19 page summaries the GLAM institutions that have closed.
From NZ? New Zealand GLAMs and COVID-19
US public libraries? Also Library Technology Guide - Public Library responses to COVID-19
Which US #publiclibraries are closed bc of #covid19? @msauers & Julie Erickson created SS + Map to track
✅Map https://t.co/FTPPCWXMUT
✅Submit new+updates: https://t.co/PyYFxYCE3z.
✅Live results: https://t.co/b47UCFw23m #covidlib #covidlibrary (submission snapshot of lower 48) pic.twitter.com/KZZtwj0vlu— Heather Braum (@hbraum) March 16, 2020
IFLA - COVID-19 and the Global Library Field
No doubt the information here can be useful for library directors who are thinking of strategies to cope with COVID-19.
For example, curious about how libraries are handing the print collection when closed?
Clearly though if your academic library isn't close yet, you should be thinking of strategies and policies in the not unlikely event you will be asked to close for a period of 2 weeks or more.
Still I have no doubts, academic libraries should do fairly well. As Lisa mentions, a huge part of the academic library service has been digital for the last decade or more.
This is the message. Academic libs have been building up a parallel digital services, content, and infrastructure for two decades. We are not moving online from in-person. We are online. We're just suspending in-person for now.
We've got this. Indiv+cooperatively. #CovidLibrary https://t.co/esCf5nMJRX
— Lisa Janicke Hinchliffe (@lisalibrarian) March 17, 2020
Resources on working in a COVID-19 world
Do you use exlibris products? - Best Practices for the Ex Libris Community During COVID-19
How to do IL online...
See also Other online resources for IL from ACRL
Managing Remote Conference Presenters with Zoom
Also don't miss out Ithaka's COVID-19 - Resources for Higher Education and Academic Libraries
Opening up of resources by Publishers, Vendors and more
As the seriousness of the COVID-19 began to dawn on everyone, corporations including publishers and library vendors began to respond by opening up services and resources at least temporarily, while others like Proquest/Exlibris scaled back on planned system changes.
This includes not just journal articles on coronavirus but also things like steaming content, ebooks (e.g. free from Vitalsource , while Proquest ebooks get unlimited ebook accesses though seems to be US/Canada specific?), live meeting software and more (e.g LibraryThing)
The following curated lists are worth following.
Infodocket - COVID-19: Special Offers, Events, News, and Updates From Library Vendors, Publishers, Online Services, and Others
Myloft's Free resources on COVID-19 - Curated list of open access content on Coronavirus related research
Free Tools for Online Teaching and Learning During School Closures
Google Scholar home page'
The Internet Archive has announced the launch of a National Emergency Library (NEL). This is not without controversy.
Datasets and analysis on COVID-19 articles
In terms of Journal articles, over 30 publishers have committed to making COVID-19 articles in Pubmed.
In particular, publishers like Elsevier are not just giving the full text in PDF but also the XMLs, which makes text data mining a lot easier (though there are doubts on the license)
You can now download a corpus of all Elsevier published research on COVID-19 + related viruses for full text and data mining, re-use and analyses for as long as needed. Incorporates 20,000+ pieces of Elsevier content, machine-readable and free: https://t.co/gbc5bSbMZb
— Elsevier (@ElsevierConnect) March 20, 2020
Here are some datasets that are worth looking at if you want to poke around at articles on COVID-19/Coronavirus
Academic Data Science alliance - COVID-19 Data Science Resources - very comprehensive.
COVID-19 Open Research Dataset (CORD-19) - 44k articles, 29k full text as at 20 March
Covid Search - index of COVID-19 Open Research Dataset (CORD-19)
Dimensions COVID-19 publications, datasets and clinical trials
Lens.org COVID-19 Dataset - patents on Cornavirus (6,896) and Scholarly works cited (18k+)
The COVID-19 Open research dataset is probably the most comprehensive, and while Allen Institute for AI which supports Semantic Scholar is leading this (also see Semantic Scholar feed on COVID-19) but it also partners with other big players like Microsoft and draws on Microoft academic graph , but one should also consider looking at the totally free Lens.org dataset.
According to Chaomei Chen, author of the powerful Science mapping tool, Citespace, Lens has quite a few virtues.
Lens data is far more comprehensive than WHO's collection. Lens: 59,106 vs WHO: 1,968. Many pieces shown here are missing in the WHO's collection.
— Chaomei Chen (@CiteSpace) March 18, 2020
In particular, Lens has one of the largest collection of patents (free) and connects it with the Scholarly papers, allowing you to visualize works cited by patents
A network overlay of a sub-network of articles that have been cited in patents based on Works Cited By Patents in https://t.co/ekrTbZaim1. pic.twitter.com/2hG8dsraDs
— Chaomei Chen (@CiteSpace) March 19, 2020
#CoronavirusResearchLandscape
How patents cited the literature of coronavirus research (1975-2017). Data Source: https://t.co/eQJPPAWOYz Patents: 12,665; g-index (k=50), LRF=-1, MLPN=10, LBY=10, e=0. Q=0.8882.https://t.co/Bo9F506WiQ pic.twitter.com/FYLWTLiNs5— Chaomei Chen (@CiteSpace) March 20, 2020
Talking about visualizing and analysis of COVID-19 data, here are some projects that aim to work on analyzing COVID-19 raw data (including metadata full text of articles) - e.g. using scrapping of data, NLP, bibliometric techniques, clustering of literature etc.
As mentioned earlier the COVID-19 Open research dataset has become the standard dataset data scientists are working on. The following is a very useful overview of the dataset and how people are usng it.
COVID-19 Open research dataset
Kaggle COVID-19 Open Research Dataset Challenge (CORD-19), TREC COVID on CORD-19
Chaomei Chen's Coronavirus Research Landscape - visualization using Citespace
Does the COVID-19 crisis expose problems in the Scholarly Communication system?
Though publishers seem to be doing their part in making data open, predictably critics of Elsevier and co are not appeased.
Publishers make 1000s of research papers on #coronavirus freely available to all.
Imagine if they had done this before the outbreak. Imagine if they did this for all public health issues.
Imagine if life-saving knowledge was a public good.https://t.co/U7QqhYtbgy #COVId19
— Jon Tennant (@Protohedgehog) March 19, 2020
As noted in "The Coronavirus (COVID-19) outbreak highlights serious deficiencies in scholarly communication" , experts sent to Wuhan at the end of 2019, analysed the data and chose to submit in The Lancet and NEJM, These papers were published on Jan 24 and 29, which is a very quick around time.
The problem is this.
Some experts had access to the first 41 confirmed cases but were reluctant to share data before publication in a prestigious medical journal, according to doctors and scientists. One doctor repeatedly asked for more clinical details and was brushed off.https://t.co/irnZhERXYa
— Jonathan Cheng (@JChengWSJ) March 7, 2020
Clearly this is a problem!
What is the alternative? Given speed is of essence, preprints deposited in preprint servers like bioRxiv, medRxiv have come into prominence.
The speed and the amount of research going into #COVID19 is unprecedented & impressive for a virus we knew nothing about 3 months ago ! To date 672 articles were posted on #bioRxiv and #medRxiv and over 120 human clinical trials are underway according to https://t.co/jLyZgBGt6d. pic.twitter.com/3C0AiKMidy
— Gaetan Burgio (@GaetanBurgio) March 20, 2020
Demo your reviewing skills by posting comments on COVID-19 bioRxiv/medRxiv preprints. Readers will benefit, it will help you learn to review, and will demonstrate your ability to write/critique <- good for other career choiceshttps://t.co/pqlcAHqibR https://t.co/Sw3u5kNeiF
— Richard Sever Ⓤ (@cshperspectives) March 13, 2020
See also- Outbreak Science Rapid PREreview
So how sound are the 500+ preprints up there? Who knows, but we do know for sure that one preprint posted on bioRxiv , titled “Uncanny similarity of unique inserts in the 2019-nCoV spike protein to HIV-1 gp120 and Gag” was retracted.
While the linked article argues this is good, because the error was quickly found, others predictably felt this was a fundamental problem with non-peer reviewed preprints and so the debate on preprints continues.
What can we do to mitigate this issue?
Our friends at scite have started to include preprints and articles on COVID19 into the database, so we can see if preprints are being cited and the way they are cited.
We added 19,696 citation statements to https://t.co/u6CTfBjL6y yesterday from 994 preprints/articles on #COVID19. This makes it easy to see how these articles are being cited and if there is supporting or contradicting evidence. Here's an example https://t.co/nLRWHoZyZ6 pic.twitter.com/8ONQ9bDntF
— scite (@scite) March 18, 2020
They show it handles the earlier retracted preprint beautifully.
Remember the #preprint "Uncanny similarity of unique inserts in the 2019-nCoV spike protein to HIV-1 gp120 and Gag"? It's since received two contradicting citations: https://t.co/RhO9R1sfTT pic.twitter.com/nRlerPKp1m
— scite (@scite) March 18, 2020
If you are reading preprints, you can use scite browser extension, which combines the power of scite and Scholarcy, particularly Scholarcy's preprint health check.
Are you writing or reading a #preprint on #COVIDー19? We've teamed up with @scholarcy to make it easy to check citation contexts of each of your references for free. "scite before you cite!" https://t.co/xo2uM241WH pic.twitter.com/ftZZ8Dlm8R
— scite (@scite) March 16, 2020
According to scite, they are in discussions with manuscript submission systems to include this functionality.
Another nice tool is using 2D search to check possible errors in advanced nested boolean searches for detecting papers on COVID-19 (for papers prior to the confirmation of the official name)
Visualizing the COVID-19 search string as proposed on the #expertsearching list. I wonder if those duplicates were intentional? https://t.co/vs7gBajP07 pic.twitter.com/RWNrn2GksL
— 2Dsearch: think outside the search box (@2d_search) March 19, 2020
Conclusion
There is supposedly a Chinese saying/curse, that goes "May you live in interesting times", though as a Chinese speaking person, I can assure you no such saying or curse exists.
In any case, we do live in such times. Thanks for reading, and stay safe out there!