The main article may not contain all the information needed to interpret a peptide experiment. Supplementary methods, source-data tables and repository records can answer different questions, and they may be stored separately. Finding them requires following the article’s specific references rather than assuming that one downloaded PDF contains everything.
Start with the article record and its file references
Check the publisher’s article page as well as the PDF. Look for supporting information, supplementary material, source data and data-availability sections, using the labels present on that record.
PMC’s user guide describes an Associated Data section that gathers available data citations, availability statements and supplementary materials. It appears when the article includes one or more of those features.NLM — PMC User Guide (opens in a new tab)
Follow the exact file or accession cited beside the result you are investigating. A reference to Supplementary Table S4 is more useful than an unspecific assumption that the data must be somewhere in the download folder.
Distinguish methods, summaries and underlying data
| Item | What it may provide |
|---|---|
| Extra methods | Additional description of the experiment or analysis |
| Summary table | Processed results or selected measurements |
| Source-data file | Values underlying particular figures or analyses |
| Repository record | Dataset metadata, files and access conditions |
| Availability statement | Where data can be obtained, or why access is restricted |
In an original hypothetical paper, the main figure displays a mean binding response, Supplementary Table S2 lists individual processed responses and a repository contains instrument output. These are linked evidence layers, not interchangeable copies of the same information.
The processed table can help reproduce the plotted mean without exposing every preprocessing decision. The instrument files may provide more detail but still need sample labels, metadata and analysis code to be interpretable.
Avoid calling a file raw data solely because its filename contains raw. Inspect its contents and documentation to identify what transformations have already occurred.
Verify that the supporting material belongs to this version
Compare the title, author list, DOI or other parent identifier with the article you are reading. Similar filenames from related studies can be easy to confuse, especially when files are downloaded separately.
Check version dates and correction notices where available. A revised main figure may have an accompanying replacement source-data file, while an older local copy still shows the earlier values.
Record the figure or table each file supports and the relevant sheet or column. In the hypothetical binding example, a spreadsheet containing a different endpoint does not validate the displayed binding calculation simply because both came from the same paper.
Also inspect legends and units. A column labelled response may be normalised, background corrected or expressed relative to a control; the numerical entries alone do not tell you which.
Record what you could and could not inspect
An accessible repository landing page does not guarantee that every associated file is openly downloadable. Follow the stated access conditions and distinguish metadata inspected from data actually examined.
If a file is missing, a link is broken or the statement requires a request, record the exact limitation. Do not fill the gap with an assumed sample size, preparation condition or analysis step.
A useful reading note identifies the parent article and version, supporting item, location, contents checked and unresolved question. It should be possible to return directly to the file that supports an interpretation.
Finding additional material is the beginning of verification. Whether it resolves the scientific question depends on its completeness, relevance and connection to the reported result, not simply on the presence of a supplementary-files button.
Sources and further detail
- NLM — PMC User Guide (opens in a new tab)
Article navigation and Associated Data guidance read. The binding-paper example and file map are hypothetical; no specific dataset is claimed to have been inspected.
Sources checked 19 September 2026. Worked examples are illustrative unless a supplied report is explicitly identified. This article has not undergone independent scientific peer review.