RIS file format specification¶

The RIS data format is a tag based file format containing one or more citations. A sample file is shown here:

TY  - JOUR
AU  - Spitz, François
AU  - Furlong, Eileen E. M.
PY  - 2012
DA  - 2012/09/01
TI  - Transcription factors: from enhancer binding to developmental control
JO  - Nature Reviews Genetics
SP  - 613
EP  - 626
VL  - 13
IS  - 9
AB  - Genetic studies during the past decades have revealed tightly regulated transcriptional networks that control robust developmental programs. However, this deterministic view of development contrasts with recent genomic studies that suggest that the multiple steps of transcription are by themselves rather leaky.Although any single transcription factor (TF) can typically bind to thousands of sites throughout the genome, cis-regulatory activity at enhancers requires the concerted action of multiple binding events (which can be homotypic or heterotypic).TFs bind to enhancers in a combinatorial manner, which is facilitated through direct and indirect cooperative mechanisms. The combinatorial nature of enhancer occupancy allows genes to be regulated in complex patterns in both space and time.DNA and proteins can act together as a scaffold to cooperatively recruit TFs to enhancers.Enhancers undergo progressive changes during development, in which their occupancy by TFs, and the position of nucleosomes with or without post-translational modifications, reflect their inactive, poised or active state.Pioneer TFs recruit chromatin-remodelling factors to reposition nucleosomes, thus facilitating the occupancy of other TFs at subsequent developmental stages.Enhancer priming can occur through many different mechanisms. For example, some TF binding events may serve as placeholders to prevent nucleosome repositioning.Partially redundant enhancers can act to buffer changes in environmental conditions to ensure robust developmental progression.Multiple elements regulating the same gene seem to assemble in a three-dimensional structure with the promoter, and can have synergistic or repressive influences on each other.
SN  - 1471-0064
UR  - https://doi.org/10.1038/nrg3207
DO  - 10.1038/nrg3207
ID  - Spitz2012
ER  -

Tag format¶

Tags are six-character long, with the following structure:

XXss-s

where X represents and alphanumeric character, s a space, and - is a dash.

Each new tag must start in a new line. However, the content from a tag can extend to multiple lines. Tags can be repeated, for instance to describe multiple authors or keywords in a reference.

Bibliographic record format¶

A bibliographic record must start with a TY tag followed by the publication type. Publication types in the standard implementation include:

ABST, ADVS, ART, BILL, BOOK, CASE, CHAP, COMP,
CONF, CTLG, DATA, ELEC, GEN, HEAR, ICOMM, INPR, JFULL,
JOUR, MAP, MGZN, MPCT, MUSIC, NEWS, PAMP, PAT, PCOMM, RPTR,
SER, SLIDE, SOUND, STAT, UNBIL, UNPB, VIDEO

A record must end with an ER tag. The rest of the tags can be in any order.

Extensions to the standard¶

In addition to the publication types listed above, other publication types that can be found include:

CLSWK, EBOOK, ECHAP, EJOUR

Bibtex-compatible publication types¶

The set of publication types covered by the RIS format is larger than that of BibTeX. The subset of types that have a clear BibTeX counterpart include:

BOOK, CHAP, CLSWK, CONF, CPAPER, EBOOK, ECHAP, EJOUR,
ELEC, GEN, JOUR, MGZN, RPRT, SER, THES, UNPB

Standard tags¶

A list of the standard tags with basic definitions are provided here:

ris_tags_ref = {
    'TY': "Type of reference - first tag in a record",
    'A1': "Primary Authors - usually mapped to AU",
    'A2': "Secondary Authors - normally used for Editors",
    'A3': "Tertiary Authors",
    'A4': "Subsidiary Authors",
    'AB': "Notes - usually used for abstract and mapped to N2",
    'AD': "Author Address",
    'AN': "Accession Number",
    'AU': "Author",
    'AV': "Location in Archives",
    'BT': "This field maps to T2 except for Whole Book and Unpublished Work references",
    'C1': "Custom 1",
    'C2': "Custom 2",
    'C3': "Custom 3",
    'C4': "Custom 4",
    'C5': 'Custom 5',
    'C6': "Custom 6 - As of 2023 some references in WoS use this for Year + Month info",
    'C7': "Custom 7",
    'C8': 'Custom 8',
    'CA': "Caption",
    'CN': "Call Number",
    'CP': "Misc",
    'CT': "Title of unpublished reference",
    'CY': "Place Published",
    'DA': "Date",
    'DB': "Name of Database",
    'DO': "DOI",
    'DP': "Database Provider",
    'ED': "Editor",
    'EP': "End Page",
    'ET': "Edition",
    'ID': "Reference ID",
    'IS': "Issue number",
    'J1': "Periodical name: user abbreviation",
    'J2': "Alternate Title - Journal name for journals",
    'JA': "Periodical name: standard abbreviation",
    'JF': "Journal/Periodical name: full format",
    'JO': "Journal/Periodical name: full format",
    'KW': "Keywords",
    'L1': "Link to PDF",
    'L2': "Link to Full-text.",
    'L3': "Related Records",
    'L4': "Image(s)",
    'LA': "Language",
    'LB': "Label",
    'LK': "Website Link",
    'M1': "Number",
    'M2': "Miscellaneous 2",
    'M3': "Type of Work",
    'N1': "Notes",
    'N2': "Abstract",
    'NV': "Number of Volumes",
    'OP': "Original Publication",
    'PB': "Publisher",
    'PP': "Publishing Place",
    'PY': "Publication year",
    'RI': "Reviewed Item",
    'RN': "Research Notes",
    'RP': "Reprint Edition",
    'SE': "Section",
    'SN': "ISBN/ISSN",
    'SP': "Start Page",
    'ST': "Short Title",
    'T1': "Primary Title",
    'T2': "Secondary Title (journal title, if applicable)",
    'T3': "Title series",
    'TA': "Translated Author",
    'TI': "Title",
    'TT': "Translated Title",
    'U1': "User definable 1",
    'U3': "User definable 3",
    'U4': "User definable 4",
    'U5': "User definable 5",
    'UR': "URL",
    'VL': "Volume number",
    'VO': "Published Standard number",
    'Y1': "Primary Date",
    'Y2': "Access Date",
    'ER': "End tag in a record"
}

This list is included in gris for reference purposes.

Specific fields information¶

One of the challenges of the RIS format is that there is not one way of describing key fields used for references. In fact, both publishers and digital libraries use different tags for the same fields, including in some cases their own extensions to the standard.

Titles¶

The tags: T1, TI, CT, and BT are assigned to the primary title of the record, with BT used only for Whole Book and Unpublished work references.

The tags: T2 and BT are used for the secondary titles. Some journals export the Journal name using T2.

Authors¶

The tags: A1 and AU are used for the primary authors of a reference, with each tag containing the name of only one author. The author name is encoded as follows:

LastName, FirstName, Suffix

Two other tages, A2 and ED, are used to list the names of Editors.

Publication date¶

The tags Y1 and PY are used to encode the publication date information, which should be in the format:

YYYY/MM/DD/extra info

As of March 2023 some articles exported from WebOfScience use the C6 tag to describe the month and year of publication as:

Month YYYY

It is unclear why is this the case and why only some of the publications seem to be affected

Abstracts¶

Abstracts are captured by either the N2 or AB tags. Sometimes both.

Journal titles¶

The standard establishes JF and JO as the tags with the periodical name, with JA used to specify the standard journal abbreviation and J1 and J2 used for user-defined abbreviations. In practice, though, many sources use T2 for the full journal name.

There is a significan lack of consistency on the use of these tags among publishers. For instance, some use JO for the abbreviated journal name instead of the full name, as described in the specification.

Volume, Issues, and Pages¶

VL and IS are used for volumes and issues. Start and end page are encoded with the SP and EP tags, respectively.