Science requires accuracy. Even if the research path is correct on the wrong basis, the result can not be accurate. In the field of genome research involving human life, a seemingly minor error may also cause unbearable consequences for human beings. I'm afraid of the following news!
In nearly 12000 papers on human genetics, more than 700 studies contained errors in DNA or RNA sequences. A report on the British website nature exposed the incident, and scientists warned that it must be vigilant.
A computer-aided analysis of nearly 12000 papers on human genetics found that papers on human gene function are full of "shocking" errors, and the proportion of nucleotide sequence errors in papers is unacceptably high. This represents a major crisis for the field of genomics aimed at transforming basic research into patient treatment, because this transformation depends very much on a reliable description of gene function.
In fact, incorrectly identified nucleotide sequences represent a "double hazard" - incorrect sequences themselves and their related results may mislead ongoing experiments and future research directions.
It may be accidental or academic fraud
According to the website of nature, the analysis found that in nearly 12000 papers on human genetics, more than 700 studies contained wrong DNA or RNA sequences. Jennifer Brian, the research leader and cancer expert at the University of Sydney, Australia, has been committed to finding major errors in genetic research since 2015. Brian believes that this proportion must be vigilant because it shows that part of human genome research is unreliable.
Researchers believe that these mistakes may be accidental, but there may also be academic fraud.
Brian first found the problem in five research papers. These papers involve a very common experiment - using a small piece of DNA to inactivate genes in cancer cells, and the nucleotide sequence used in the experiment is wrong. In addition, these five papers also use very similar language expressions and data.
Brian suspected that they might come from a "paper factory", that is, papers delivered by a third-party company according to the "order".
At present, four of the five papers have been withdrawn. But Brian's eyes began to notice other papers with similar errors.
The error rate is unacceptably high
In 2017, Brian worked with computer scientists at the University of Grenoble in France to create a software called "seek & BLASTN", which can identify and mark potential errors in research.
For example, the software can extract short nucleotide sequences from papers and compare them with the data in the open nucleotide database "BLASTN" to check whether they match the expected human genome. Next, the researchers will manually check each marked mismatch.
Using the software, the research team screened the relevant papers published in the journals of gene and oncology report, including 7400 original papers published in gene from 2007 to 2018 and 3800 open access papers published in oncology report from 2014 to 2018.
After manual screening, the researchers found that about 12% of the nucleotide sequences in the oncology report had problems《 2% of the papers on genes have the above problems.
In addition, Brian and the research team also screened the research papers in the sub field of cancer genetics that had problems before. The results showed that in about 600 relevant papers, more than 25% of the papers had wrong nucleotide sequences.
The researchers said that the proportion of papers with wrong nucleotide sequence found this time was unacceptably high, and a total of 712 papers (about 6% of the total screening) had wrong nucleotide sequence. These studies have been published in 78 journals and have been cited more than 17000 times.
Is it just the tip of the iceberg?
Nucleotide sequence refers to the sequence of bases in DNA or RNA, which is the basis of a series of molecular genetic techniques. These technologies have been applied in hundreds of thousands of research publications.
However, according to the researchers' estimation, about a quarter of the problem papers may mislead or distract the future development of human disease treatment. Therefore, the researchers call for immediate and urgent measures to solve the problem of unreliable gene function papers in the literature.
The team has sent emails to all relevant journals or publishers that can find the editor's contact information, and some of them replied that they will investigate these problematic papers.
But at the same time, the researchers also hope that this research can inspire other researchers and journals to find more problematic papers on human gene function, because they still have greater concerns - the wrong papers found this time are only the tip of the iceberg