---
name: "phrase-search"
section: "1"
platform: "linux"
package: "ncbi-entrez-direct"
version: "19.0.20230216+dfsg-2+b1"
release: "Debian 12 (Bookworm)"
platforms: ["linux"]
url: "https://new.carta.tech/phrase-search"
---

# phrase-search(1)

search an indexed local NCBI PubMed archive

## NAME

phrase-search - search an indexed local NCBI PubMed archive

## SYNOPSIS

phrase-search \[**-h**\|**-help**\|**--help**\] \[**-path** *dir*\] **-count**\|**-counts**\|**-query**\|**-filter**\|**-link**\|**-exact**\|**-title**\|**-words**\|**-pairs**\|**-fields**\|**-terms**\|**-totals** *query...*\|**-convert** *dbas* *type* *type* \[*key*\]\|**-lookup** *tag* *theme...*

## DESCRIPTION

phrase-search searches an indexed local NCBI PubMed archive (as prepared by [**index-pubmed**(1)](/index-pubmed/1.md)).

## OPTIONS

- **-h**\|**-help**\|**--help**

  Print usage information.

- **-path** *dir*

  Search the local archive in *dir*.

- **-count**

  Print terms and counts, merging wildcards.

- **-counts**

  Expand wildcards; print individual term counts.

- **-query**

  Search on words or phrases in Boolean formulas.

- **-filter**

  Filter piped-in output from a previous (quite possibly remote) query.

- **-exact**

  Strict search for article round-tripping.

- **-title**

  Exact search limited to indexed title field.

- **-link**

  Follow links by the given tag (*e.g.* **CITED**), merging any duplicates.

- **-words**

  Combine searches for individual informative words.

- **-pairs**

  Combine searches for pairs of adjacent informative words.

- **-fields**\|**-terms**

  Interpret the query as a field name, and list that field's known terms. With an empty query, list supported fields.

- **-totals**

  Like **-fields**, but with term counts.

- *query...*

  What to search for in the above modes; may contain standard Boolean operators **AND**, **OR**, and **NOT**, using parentheses as needed for grouping.

- **-convert** *dbas* *type* *type \[key\]*

  *Convert identifiers in the specified database* *from one type to another,* *accepting either a single identifier on the command line* *or a newline-delimited list on standard input.* *The database may be any of*

  - **mesh**

    *Medline subject headings in general*

  - **chem**

    *MeSH chemicals*

  - **disz**

    *MeSH diseases*

  - **gene**

    *Genes*

  *and the type may be any of*

  - **Name**

    *official name*

  - **Code**

    *internal identifier*

  - **Tree**

    *path within MeSH tree (except for* **gene***)*

  - **Desc**

    *description (only for* **gene***)*

  - **Term**

    *synonym (except for* **gene***)*

- **-lookup** *tag theme...*

  *Look up cross references between internal identifiers* *read from standard input,* *where tag identifies a pair of databases* *(e.g.,* **chdi** *for chemicals and diseases)* *and theme... are theme codes of particular interest* *(with* **p** *and* **m** *standing in for* **+** *and* **-** *respectively).*

## ENVIRONMENT

- **EDIRECT\_PUBMED\_MASTER**

  *Local archive directory to use* *in the absence of* **-path***/***-master***.* *Expected to hold an absolute path;* *mandatory when not supplying a path on the command line.*

## SEE ALSO

[**cit2pmid**(1)](/cit2pmid/1.md), [**download-ncbi-data**(1)](/download-ncbi-data/1.md), [**edict**(1)](/edict/1.md), [**fetch-pubmed**(1)](/fetch-pubmed/1.md), [**filter-stop-words**(1)](/filter-stop-words/1.md), [**index-pubmed**(1)](/index-pubmed/1.md), [**rchive**(1)](/rchive/1.md), [**theme-aliases**(1)](/theme-aliases/1.md), [**word-at-a-time**(1)](/word-at-a-time/1.md), [**xtract**(1)](/xtract/1.md).