>jane*

Analyze

PDF

Keep the paper with the data. A protocol, a supplementary file or the study your constructs came from opens as a real tab beside the sequences it describes, page by page, with zoom that re-renders rather than stretching — so the figure you are squinting at stays sharp.

A real published PLOS ONE study as a PDF, beside the supplementary workbooks it published — every sheet selectable, dates read as dates rather than serial numbers

01

What it does

  1. 01A PDF opens as its own tab on the stage, beside the sequences and data it describes
  2. 02Page-by-page navigation over the whole document
  3. 03A thumbnail strip of every page beside the document, so you can jump by what a page looks like
  4. 04A Section control built from the document’s own bookmarks, which also names the section you are currently in
  5. 05Zoom re-renders the page at that size, so figures and small type stay sharp rather than being stretched
  6. 06Rotate for a landscape figure, or a page scanned sideways
  7. 07The document’s own title, author, PDF version and producing application are shown when it declares them
  8. 08Pages are parsed and rasterised in a worker, so a long document does not freeze the rest of the workbench
  9. 09The same viewer draws the file manager preview and the stage tab
  10. 10The page’s text is selectable and copyable, over the rendered drawing
  11. 11PDF text is searched with the rest of the project — hits report the page
02

What it runs on

Rendering is done by pdf.js, the reader Mozilla builds and ships in Firefox, used as a library rather than reimplemented. One page is rendered at full size — the one you are reading — while the strip’s thumbnails are drawn one after another behind it, with no cap on how many. The page count and metadata come from the document’s cross-reference table, which is cheap even for a large scan.

Everything runs client-side. Files are parsed in the browser and projects are saved on this machine, so an unpublished sequence is never uploaded.

03

Questions

Is my document uploaded anywhere?
No. It is rendered in the browser from the copy stored in your project.
Can I select or copy the text?
Yes. An invisible, correctly positioned copy of the page’s text sits over the drawing, so you can select and copy a gene name, an accession or a primer sequence straight out of the paper.
Can I find a word across my whole project?
Yes — a PDF’s text is searched alongside your sequences and tables, and a hit reports the page it is on rather than a character offset, because a PDF stores positioned glyphs and reading order is a layout property. The first 40 pages of a document are searched; a longer one says so.
Why does zooming stay sharp?
Because zoom re-renders the page at the new size rather than scaling a bitmap. That is the difference that matters when the reason you zoomed was to read a gel lane or an axis label.
Can I jump straight to a section?
If the document ships bookmarks, yes — a Section control lists them in the document’s own order and nesting, and it also names the section the page you are on belongs to. Most journal PDFs ship none, and then the control is not shown rather than shown empty. The bundled 223-page reference manual has 142 of them.
Is there a page thumbnail view?
Yes — a strip down the left draws every page small, in order, and clicking one jumps to it. It fills in progressively rather than making you wait, and it is not capped at some page number.
How large a PDF can it open?
Up to 128 MB. Anything larger is refused with the size stated rather than loaded.

Also searched as

read a paper next to my data · pdf viewer in the browser · open a protocol pdf · keep the paper with the plasmid · lab notebook pdf viewer · open supplementary pdf online · view a pdf without leaving the app