Free online NBIB to CSV converter

PubMed and other NCBI services export citation records as NBIB, while spreadsheets, databases, and analysis tools commonly require CSV. Converting the file makes citations easier to sort, filter, edit, and import without manually copying fields.

What the NBIB format is

NBIB is a plain-text bibliographic format based on tagged MEDLINE records. PubMed users and researchers in medicine, life sciences, and other academic fields use it to exchange references with citation-management software. A record contains tags such as PMID- for the PubMed identifier, TI - for the title, AU - for an author, DP - for the publication date, and AB - for the abstract.

PubMed creates .nbib files through Send to → Citation manager. The format supports repeated tags and continuation lines, so one citation can contain multiple authors, MeSH terms, article identifiers, and lines of abstract text. A text editor can display an NBIB file, but its tagged, multiline structure is inconvenient for tabular editing.

What the CSV format is

CSV is a delimited text format in which each row represents one record and each column represents one field. A citation CSV can have columns for PMID, title, authors, journal, publication date, abstract, DOI, and publication type.

CSV is supported by Microsoft Excel, LibreOffice Calc, Google Sheets, databases, and programming libraries. It is useful for filtering, deduplication, reporting, and importing citations into systems that do not read NBIB. CSV is not a fixed schema: the converter must decide which NBIB tags become columns, and spreadsheet programs may use a semicolon delimiter in some regional settings.

Convert NBIB to CSV online

You can convert the file on 101convert.com for free without installing software. Upload the .nbib file, choose CSV as the output format, start the conversion, and download the resulting .csv file. Open the result and inspect the column names and several records before using it for analysis.

Do not upload unpublished or confidential bibliographic data to an online service unless its handling and deletion policies are acceptable. For private or repeatable processing, use a local script instead.

Convert NBIB to CSV with Python

Biopython can parse NBIB/MEDLINE tags, while Python's standard csv module handles commas, quotation marks, and line breaks correctly.

1. Install Biopython with python -m pip install biopython.
2. Save the following code as nbib_to_csv.py.
3. Run python nbib_to_csv.py input.nbib output.csv.

import csv
import sys
from Bio import Medline

fields = [
    ("PMID", "PMID"),
    ("Title", "TI"),
    ("Authors", "AU"),
    ("Journal", "JT"),
    ("Publication date", "DP"),
    ("Abstract", "AB"),
    ("DOI", "LID"),
    ("Article ID", "AID"),
    ("Publication types", "PT"),
    ("MeSH terms", "MH"),
]

def value(record, tag):
    item = record.get(tag, "")
    if isinstance(item, list):
        return "; ".join(str(x) for x in item)
    return str(item)

with open(sys.argv[1], encoding="utf-8") as source, open(
    sys.argv[2], "w", newline="", encoding="utf-8-sig"
) as target:
    writer = csv.writer(target)
    writer.writerow([name for name, tag in fields])
    for record in Medline.parse(source):
        writer.writerow([value(record, tag) for name, tag in fields])

The utf-8-sig encoding helps older Excel versions recognize non-ASCII characters. Add fields such as TA for the journal abbreviation, FAU for full author names, or PMC for a repository identifier when those tags exist in the source.

Quality and compatibility checks

NBIB records do not all contain the same tags. Missing tags become empty CSV cells, and repeated tags must be combined rather than reduced to the first value. The script joins repeated authors, MeSH terms, and identifiers with semicolons; CSV quoting preserves commas, quotation marks, and line breaks in abstracts and titles.

A DOI may be stored in LID with a suffix such as [doi], so remove that suffix only if the destination system requires a bare DOI. Check PMID, DOI, author, date, abstract, and row counts after conversion. CSV does not inherently preserve NBIB tags, field order, repeated-field structure, or distinctions between different identifier types unless the export defines separate columns for them.