Retractions in Web of Science and OpenAlex

The Summa Cum Fraude team dealt with retractions as part of a university project. As the person responsible for the open access repository at my university and member of this team, I am particularly interested in whether and how articles that have been retracted in a journal are still available as regular (non-retracted) articles in Open Access repositories. In the absence of a workflow that queries the CrossRef API against retractions in our repository, I did a quick and dirty (!) check for retractions in Web of Science and OpenAlex to see if the Web of Science or the recently much-praised OpenAlex would provide me with better information on retractions of articles published at my university.

So I searched for publications from my university (Saarland University, Germany) with the status “retracted” in both databases. The following lists are not meant to be blaming, rather I learnt in Summa cum Fraude that retractions are a good thing and by no means per se the result of intentional misconduct.

Web of Science

The Web of Science reported only three retracted papers:

1. Chemical Composition and Biological Evaluation of Typha domingensis Pers. to Ameliorate Health Pathologies: In Vitro and In Silico Approaches
   Source: Biomed Research International
   DOI: 10.1155/2022/8010395
   Published: 2022  
2. Photoactivatable Hsp47: A Tool to Regulate Collagen Secretion and Assembly
   Source: Advanced Science
   DOI: 10.1002/advs.201801982  
   Published: 2019  
3. CXCR2 mediates NADPH oxidase-independent neutrophil extracellular trap formation in cystic fibrosis airway inflammation
   Source: Nature Medicine
   DOI: 10.1038/nm.2209
   Published: 2010 

OpenAlex

OpenAlex in turn provided me seven retracted works:

1. Functional surface microstructures inspired by nature – From adhesion and wetting principles to sustainable new devices
   Source: Progress in Materials Science
   DOI: 10.1016/j.pmatsci.2021.100778
   Published: 2021
   Note: This article was not retracted due to scientific reasons, but withdrawn, see also the publisher information: “This article has been withdrawn due to the incorrect Volume assignment.”
2. Long non-coding RNA DIO3OS binds to microRNA-130b to restore radiosensitivity in esophageal squamous cell carcinoma by upregulating PAX9
   Source: Cancer Gene Therapy
   DOI: 10.1038/s41417-021-00344-2
   Published: 2021
3. Narratives of vicarious experience in professional and workplace contexts: Introduction to the special issue
   Source: Journal of Pragmatics
   DOI: 10.1016/j.pragma.2019.12.001
   Published: 2020
   Note: This article was not retracted due to scientific reasons, but withdrawn, see also the publisher information: “The Publisher regrets that this article is an accidental duplication of an article that has already been published in Journal of Pragmatics, 155C (2020) 64-69, https://doi.org/10.1016/j.pragma.2019.10.010. The duplicate article has therefore been withdrawn.”
4. Photoactivatable Hsp47: A Tool to Regulate Collagen Secretion and Assembly
   Source: Advanced Science
   DOI: 10.1002/advs.201801982  
   Published: 2019  
5. Influence of variation in global sperm DNA methylation level on the expression level of protamine genes and human semen parameters
   Source: Andrologia
   DOI: 10.1111/and.13484
   Published: 2019
6. A highest stable cluster Au₅₈ (C₁) re-optimized via a density-functional tight-binding (DFTB) approach
   Source: RSC Advances
   DOI: 10.1039/c7ra13171b
   Published: 2018
7. CXCR2 mediates NADPH oxidase–independent neutrophil extracellular trap formation in cystic fibrosis airway inflammation
   Source: NATURE MEDICINE
   DOI: 10.1038/nm.2209
   Published: 2010 

Some considerations

Of the seven articles in the OpenAlex results list, two were not retractions but withdrawals. Unfortunately, the “retracted” filter in OpenAlex does not distinguish this and confuses both item types. Of the remaining five records, all were indeed retracted, just like in the Web of Science sample. Even if this test was only a kind of experiment, OpenAlex seems to be a good option for searching for retractions. Six of the seven OpenAlex records were also indexed in the Web of Science, only number three was missing – which raises the question as to why the Web of Science reported retractions worse than OpenAlex in this miniature sample. Like number three, record number one was a withdrawal and indexed in the Web of Science but not marked as a withdrawal. In principle, the Web of Science (unlike OpenAlex) distinguishes between retractions and withdrawals, and even uses further related markers such as “Expression of Concern”. In this sample, OpenAlex identified retractions better than the Web of Science, which in turn knows finer differentiations but did not make use of them in this sample.

However, since the OpenAlex list only covered two of the three records from the Web of Science list, one should consider not relying on OpenAlex alone. Since the publication missing from the OpenAlex list was not published until 2022 and was not retracted until the end of 2023, the cause could be a delayed indexing of the retraction.

As I learnt during the Summa cum Fraude project mentioned above, a differentiation of the status retracted would also be necessary, for example: Article five of the OpenAlex results has an interesting retraction note: One of the authors “has declared that he was unaware of this submission, did not approve the manuscript and disagrees with some of its scientific conclusions.” It seems that this person was not informed about the submission, strange as this is, no scientific error was criticised.