PubMed Abstract Listing Fetcher

PubMed Abstract Listing Fetcher is a small open-source Electron desktop application by Ville Langén, which helps you turn a plain-text bibliography into a single continuous HTML page of PubMed abstracts—useful when you are drafting, reviewing, or annotating a manuscript and want the abstract text beside each matched reference. Downloads (Windows and macOS installers) are on GitHub Releases (latest).

The Problem

When you already have a numbered reference list (for example Vancouver-style, one reference per line), pulling the matching abstracts from PubMed by hand is slow and repetitive. You might want:

• One readable document that follows the order of your list
• A quick way to skim or archive abstract text next to each entry
• Outputs you can share or file away (HTML, PDF, plain text) without touching your manuscript’s project folder until you choose to

The Solution

On GitHub, the project is available in two forms:

• Desktop app: use the graphical app described on this page.
• Command line: run it with Node.js (node pubmed_abstracts.js—see For Linux (and command-line users) below).

In the app, you paste references or load a .txt file, run Get abstracts & save ZIP, watch terminal-style progress, and when the run finishes you pick where to save a ZIP bundle. Work is performed under the operating system’s temp directory until you confirm a save path—nothing is written into your manuscript folders or elsewhere unless you choose to save it there.

Use Cases

Suited for:

• Dissertation preliminary examiners and other reviewers who want a compact abstract listing aligned to a reference list
• Students and researchers assembling reading notes from a bibliography
• Anyone who prefers a local-first workflow: the app runs on your machine; you decide where exports live

It is not marketed for regulated clinical decisions, legal filings, or guaranteed completeness of PubMed matches—always verify critical citations in the primary literature.

Why PubMed Abstract Listing Fetcher?

• Order-preserving: Your list drives the document, not a separate search session
• Explicit saves: Temp working files and a clear save dialog, so you are not surprised by files scattered through a project tree
• Practical bundle: HTML for reading, PDF for sharing or printing, TXT for grep-friendly archives

Turn a static reference list into a browsable abstract companion—on your own computer, at your own pace.

Key Features

• Plain-text input: Paste into the app or load a .txt file. For how each line must look, see Reference list format below.
• Sample list: Optional control to download a minimal example references.txt you can edit and replace.
• Progress feedback: Terminal-style log while each line is resolved and abstracts are fetched.
• ZIP export: pubmed_abstracts.html, pubmed_abstracts.pdf (printed from the HTML via bundled Chromium), and pubmed_abstracts.txt (plain text derived from the HTML).

How to Use (desktop app)

1. Open the app and paste your reference list, or use the file picker to load a .txt file
2. Optionally download the example references.txt to see the expected shape of each line
3. Click Get abstracts & save ZIP and wait until progress completes
4. When prompted, choose where to save the ZIP; extract it to get HTML, PDF, and TXT

Reference list format (references.txt)

• Use one reference per line; empty lines are skipped.
• Leading list numbers (for example 1. …, 2. …) are fine—you do not need to remove them before running this program; they are stripped internally when searching PubMed.
• References should look Vancouver-like (typical numbered biomedical lists):
  • authors first (often with et al.)
  • then the article title
  • then journal, year, DOI, and similar details
• The fetcher needs a clear boundary between the author block and the start of the title. Use either a full stop (period) or a colon there, for example:
  • …Lastname AB, Other CD. Title starts here…
  • …Lastname AB, et al: Title starts here…
• Other colons in the line, such as in volume/issue 14(1):179 or in doi:…, are ignored when choosing that split.

See the sample references.txt in the project files on GitHub for real lines.

Download Options

Ready-built installers are on GitHub Releases (latest):

• Windows (x64): portable .exe under Assets on the latest release. If SmartScreen warns about an unknown publisher, you may need More info → Run anyway
• macOS: .dmg under Assets—pick arm64 for Apple Silicon (M1/M2/M3/…) or the Intel/x64 build for older Macs. Drag the app to Applications; if Gatekeeper blocks the first launch, try Right-click → Open once

For Linux (and command-line users)

On Linux you can skip the Electron UI and run the engine directly with Node.js 20 or newer:

1. Download or clone the project from GitHub and cd into it
2. Place your references in references.txt in the working directory (or pass explicit input/output paths—see the README on GitHub)
3. Run:

node pubmed_abstracts.js

Defaults write pubmed_abstracts.html in the current working directory (this path is for the script, not the desktop app’s temp-and-save flow). Optionally set NCBI_API_KEY and NCBI_CONTACT_EMAIL in the environment for higher rate limits and polite identification with NCBI.

Technical Details

• Built with: Electron (desktop UI), Chromium for PDF generation, and the same JavaScript matching/fetch pipeline as pubmed_abstracts.js
• License: MIT (full text on GitHub and in-app via the LICENSE modal)
• Data handling: Fetches use NCBI E-utilities; behavior and availability follow NCBI’s policies and limits

Development & Support

The project is open source. Bug reports, feature ideas, and pull requests belong on GitHub. Questions or issues? Use the Issues tab there.