Clinical Orthopaedics and Related Research® publishes both clinical and laboratory research articles that influence how we think about the important problems that influence our specialty [9, 10]. Most articles we publish take forms that are familiar to most readers. Clinicians are comfortable with randomized trials, case series, and systematic reviews; although laboratory research is more varied, most clinicians can read and enjoy a well-presented science paper, whether it’s an engineering study about comparing loads to failure or an animal model of fracture healing. What makes these article types familiar is that readers have some degree of comfort assessing the content on the basis of common reporting standards.
But resistant problems call for new tools, and articles about new tools will feel unfamiliar at first. When our authors and readers need some additional guidance, we use this space to share the standards we seek from newer article types. In the past, we have done this for survey studies , large database studies , and product-testing studies ; we also have tried to help readers get more from what they read in network meta-analyses , no-difference studies , and clinical research more generally . As we enter the age of genome-driven personalized medicine and diagnostics, novel methods for the measurement of mRNA, microRNAs, and epigenetic markers (heritable changes in genes that do not involve changes in DNA sequence) now are available, but they can be confusing to read.
Thus, we feel it is a good time to provide for our authors and readers some background and direction on how gene analyses are handled in the research community today, particularly in terms of what CORR® will be expecting from a study that utilizes gene analyses.
The overarching theme in these kinds of functional genomics studies is that they seek to understand the individual and integrated biological function of genes and proteins and other components of the genome, particularly as they relate to the function of cells, tissues, and organisms. Examples of these technologies include Serial Analysis of Gene Expression (SAGE), RNA-Seq, and commercial in-situ synthesized microarrays (e.g. Affymetrix). The good—and challenging—aspects of these methods is that they produce massive amounts of raw data that cannot be directly published in any practical manner.
Gene expression studies, most commonly using commercial microarrays, have become of keen interest in part because they can be done fairly easily by sending biological specimens to a university core facility or commercial testing facility that then returns information on the simultaneous expression of thousands to tens of thousands of genes. The information has the potential to be valuable to basic scientists, as well as for diagnosing disease. However, the fundamental principles of good scientific research still apply; that is, association like up- or down-regulation of a gene with a particular condition such as osteoarthritis does not necessarily demonstrate causation. As Henri Poincaré expressed in his 1902 book Science and Hypothesis, “... a collection of facts is no more a science than a heap of stones is a house” .
With all of this in mind, we offer the following guidelines to authors who seek to publish studies on these topics in CORR, but we emphasize that these are equally useful to help orient readers to these study designs.
Our first guideline is that we will be asking authors who utilize microarray or other types of large-scale analysis of gene expression to demonstrate that they have deposited their full dataset into an international public repository, preferably the Gene Expression Omnibus (GEO). The GEO is provided by the National Center for Biotechnology Information, which archives and freely distributes microarray, next-generation sequencing, and other forms of high-throughput functional genomics data . The GEO provides a database for high throughput data, offers simple submission procedures and formats, and provides user-friendly mechanisms to work with gene expression profiles of interest. A comparable database like the European Bioinformatics Institute  is also acceptable as well as other more-specialized databases such as MethBase, a database of DNA methylation , as-appropriate. It is important—and has become a standard practice—to deposit data into one of these repositories because even though a publication may include tables of dozens or hundreds of genes that authors have identified as possibly relevant to their study question, there is also likely to be data on thousands of other genes that may not be of particular interest to the author but may be valuable to other investigators. (For small-scale datasets—50 to 100 targeted gene analyses—supplementary material in a spreadsheet-type file format permitting efficient extraction may be sufficient).
The Functional Genomics Data (FGED) Society  advocates for open access to genomics datasets and has devised standards for minimum information specifications for reporting data to assist in the process. Although the FGED society was originally formed to establish standards for databases derived from DNA microarray experiments, it has broadened its scope to include data generated by any functional genomics technology applied to genomic-scale studies. These standards include Minimal Information About a Microarray Experiment (MIAME) and MINimal information about a high throughput SEQ Experiment (MINSEQE) . Additional standards provide standard terminology for microarray experiments used in encoding data from microarray experiments into software tools and databases, including MicroArray Gene Expression-Object Model (MAGE-OM) , MicroArray Gene Expression-TAB-delimited (MAGE-TAB) , and MGED Ontology .
Our second guideline is that we will be expecting authors to confirm apparent differences in expression using an orthogonal method; for example, one might use quantitative PCR (qPCR) analysis  to confirm microarray results. Such confirmation is important so that the study goes beyond being a purely descriptive report of microarray data. Biological journals such as the Journal of Immunology have similar expectations .
CORR® welcomes manuscripts that report microarray data and other kinds of research involving functional genomics. These analyses are becoming increasingly important both to orthopaedic clinicians and musculoskeletal researchers. For those reasons, as noted earlier, we will require that data be deposited in an appropriate database and that studies will include the results of experiments that go beyond descriptive analyses to examine the importance to clinicians or researchers about the observations made.
1. Dobbs MB, Gebhardt MC, Gioe TJ, Manner PA, Porcher R, Rimnac CM, Wongworawat MD, Leopold SS. Editorial: How does CORR® evaluate survey studies? Clin Orthop Relat Res. 2017;475:2143–2145.
5. Grauer JN, Leopold SS. Editorial: large database studies--what they can do, what they cannot do, and which ones we will publish. Clin Orthop Relat Res. 2015;473:1537–1539.
6. Leopold SS. Editorial: Getting the most from what you read in orthopedic journals. Clin Orthop Relat Res. 2017;475:1757–1761.
7. Leopold SS. Editorial: No-difference studies make a big difference. Clin Orthop Relat Res. 2015;473:3329–3331.
8. Leopold SS. Editorial: “Pencil and paper” research? Network meta-analysis and other study designs that do not enroll patients. Clin Orthop Relat Res. 2015;473:2163–2165.
9. Leopold SS. Our enthusiasm for “related research”. Clin Orthop Relat Res. 2013;471:3069–3070.
10. Leopold SS, Swiontkowski M, Haddad F. Editorial: JBJS, The Bone & Joint Journal
, and Clinical Orthopaedics and Related Research
require prospective registration of randomized clinical trials-Why is this important? Clin Orthop Relat Res. 2017;475:1–3.
15. Poincaré P. Science and Hypothesis [in French]. New York, NY:Walter Scott Publishing Co. Ltd; 1905.
16. Qureshi M, Ivens A. A software framework for microarray and gene expression object model (MAGE-OM) array design annotation. BMC Genomics. 2008;9:133.
17. Rimnac CM, Leopold SS. Editorial: Basic science, applied science, and product testing. Clin Orthop Relat Res. 2014;472:2311–2312.
18. Stoeckert CJ Jr, Parkinson H. The MGED ontology: A framework for describing functional genomics experiments. Comp Funct Genomics. 2003;4:127–132.