iCite: Translation is a powerful web application that provides a panel of information summarizing the content of journal publications within a defined analysis group (where an analysis group can consist of a single article or a very large group of articles). Articles are assigned scores in three categories: Human, Animal, and Molecular/Cellular Biology, based on the number of Medical Subject Heading (MeSH) terms they have that fall into each of these categories. The data produced by iCite: Translation can be downloaded as a customized report from the dashboard and could be used to understand the content of articles within an analysis group. An example application for iCite: Translation might be to compare how close to human clinical applications two portfolios of articles are. iCite: Translation is limited to analyzing only articles that appear in PubMed; users upload the PubMed IDs for articles within the analysis group of interest.
iCite: Translation uses data and visualizations from the Triangle of Biomedicine. This approach uses MeSH terms from PubMed to classify papers as a combination of Human, Animal, or Molecular/Cellular Biology research. Papers are plotted on a tripartite graph, where human-oriented research is closest to the top corner of the triangle, animal oriented research is closest to the bottom right corner, and molecular/cellular biology research is closest to the lower left corner.
The following summary statistics are produced using iCite: Translation:
- Total number of articles within the analysis group (Total Pubs)
- Mean number of articles published per year (Pubs/Year)
- Average “Human” score for articles in the analysis group
- Average “Animal” score for articles in the analysis group
- Average “Molecular/Cellular” score for articles in the analysis group
- Average “Approximate Potential to Translate” score for articles in the analysis group (a machine learning-derived estimate of the likelihood that an article will be directly cited by a clinical article)
- The number of articles within the analysis group that has already been cited by a clinical article
iCite: Translation uses two triangle visualizations: a bubble plot, and a density plot.
The density plot shows the relative density of articles in an area of the triangle (this is accomplished by treating each coordinate as a point light source, applying a Gaussian blur, and visualizing in pseudocolor where ‘hotter’ colors mean higher article density).The bubble plot shows bubbles located on the triangle where papers are located. Multiple papers can have the same coordinates, so the size of the bubble indicates how many papers are there.
An example of human-oriented research
Frequency and timing of nonconvulsive status epilepticus in comatose post-cardiac arrest subjects treated with hypothermia, Neurocrit Care, 2012
An example of animal-oriented research
Force plate gait analysis in Doberman Pinschers with and without cervical spondylomyelopathy, J. Vet. Intern. Med., 2013
An example of molecular/cellular biology-oriented research
Foxp1 and lhx1 coordinate motor neuron migration with axon trajectory choice by gating Reelin signaling, PLoS Biol., 2010
An example of a mixed portfolio
The bubble plots provided by iCite: Translation are interactive. If you click on a bubble, only papers in that bubble will appear in the sortable table below the bubble plot:
Multiple bubbles can be selected with a click-and-drag gesture with the mouse. Draw a shape around the bubbles of interest. The shape is complete when a dotted line appears between the shape’s starting and ending points as shown below.
The surrounded bubbles will then be selected.
If users draw multiple shapes, bubbles will be added to the selection. To clear, press Esc key on keyboard or click the “Clear filters” button. Article-level data can be downloaded by clicking the “Export” button. Only the selected documents will be included in the download.
Note that if a paper has no MeSH terms that are categorized as Human, Animal, or Molecular/Cellular Biology (e.g. a physical chemistry paper, or a paper that lacks MeSH terms because it is not indexed in MedLine), it will appear below the triangle.
iCite: Translation also displays three histograms, with the average Human, Animal and Molecular/Cellular Biology scores in each publication year, so that trends over time can be easily identified:
As one measure of bench-to-bedside translation, iCite: Translation gives a count of the number of papers in this portfolio that have been cited by a clinical article. For primary research articles, citation in a clinical study is a strong indication that the knowledge generated by that study has transitioned into clinical research. Clinical articles are generally defined as Clinical Trials or Clinical Guidelines. Specifically, iCite: Translation uses Publication Type flags from PubMed to define clinical articles (documented on the Data help page).
In the table of article-level data, if clinical articles have cited a paper in this analysis group, a link indicating the number of clinical citations appears in the “APT” Column (described in detail below).Clicking the link will bring up a PubMed results page with the list of clinical articles that cite this publication. iCite: Translation users should note that bench-to-bedside translation takes several years. For this reason, significantly fewer papers from recent years have been cited by clinical articles, even if they eventually will be. The graph below shows the proportion of published in each year that have been cited by a clinical study, using data through 2014:
Because of the long lag from publication to usage in a clinical study, the Office of Portfolio Analysis set out to develop an approach to identify early signatures of bench-to-bedside translation. Our measure, Approximate Potential to Translate, uses the scientific community’s reaction to an article to estimate the likelihood that the knowledge from that paper will be used and cited in later clinical articles. Approximate Potential to Translate is expressed as a probability from 0.05 (no detectable signatures of translation) to 0.95 (extremely strong signatures of translation). It is generated by training a machine learning model with information describing the Human, Animal, and Molecular/Cellular scores of the article in question as well as those of the later articles that cite the paper. The machine learning system learns, based on whether the training articles were eventually cited by a clinical article, which features serve as early signatures that knowledge from a publication is moving toward clinical applications. It then uses these signatures of bench-to-bedside translation make predictions for each article in iCite.
For example, 25% of the papers that have a Approximate Potential to Translate of 0.25 are cited by clinical articles on average. Likewise, 75% of papers that have a Approximate Potential to Translate of 0.75 are cited by clinical articles on average. The fact that 25% of papers with a Approximate Potential to Translate score of 0.75 are not cited by a clinical study is not an error, but rather the intended design of this metric.
There are some options for selecting and deselecting subsets of the data based on publication year or type.
The year field has dropdown menu filters, which can be used to select a specified year range. Filters set in this way can be cleared by clicking the "Clear filters" button.
In the article-level table, there are further customization options. You can see the title, authors, journal, article type, and RCR of each paper:
Users can select and deselect individual papers with the checkbox on the left (for example, deselecting a person's middle-author papers). In addition, whole years can be selected or deselected. Selecting all or deselecting all, by clicking the check box next to the "PMID" column header, can be used to reset the individual selections. Note that other filters that have been set will remain in place. Columns are sortable, which is done by clicking the columns header. Clicking on "Year" will sort by publication year, for example, and clicking on RCR will sort by RCR.
Users can navigate the citation network by clicking the "Citing Papers" and "Referenced Papers" buttons. Clicking "Citing Papers" will open a new iCite page to view the papers that cited the ones in this analysis group, while clicking "Referenced Papers" will open a new iCite page to view the papers that were referenced by those in the analysis group.