Skip to content

Unified Medical Language (UMLS) Filtering

nbviewer

This example demonstrates how to obtain an occurrence matrix associated with a UMLS concept in a previously obtained pubmed/medline search.

Note: This example uses the database created and updated by:

  • Examples / Pubmed Search and Save
  • Exmaples / MeSH/UMLS Map

The following backends are supported for retieving the prior information:

  • MySQL
  • SQLite

Set Up

using BioMedQuery.Processes
using BioMedQuery.PubMed
using MySQL
using SQLite

results_dir = ".";
umls_concept = "Disease or Syndrome";

MySQL backend

Connecting to MySQL database that was created in pubmed_search_and_save example

host = "127.0.0.1";
mysql_usr = "root";
mysql_pswd = "";
dbname = "pubmed_obesity_2010_2012";

db_mysql = MySQL.connect(host, mysql_usr, mysql_pswd, db=dbname);

Getting the descriptor to index dictionary and the occurence matrix

@time labels2ind, occur = umls_semantic_occurrences(db_mysql, umls_concept);

Filter mesh query string : SELECT mesh FROM mesh2umls WHERE umls IN ( "Disease or Syndrome" )
-------------------------------------------------------------
Found 5 articles with valid descriptors
-------------------------------------------------------------
  0.005654 seconds (2.62 k allocations: 163.453 KiB)

Descriptor to Index Dictionary

labels2ind

Dict{String,Int64} with 5 entries:
  "Obesity"                  => 1
  "Pediatric Obesity"        => 2
  "Sleep Apnea, Obstructive" => 3
  "Metabolic Syndrome"       => 4
  "Fibromyalgia"             => 5

Output Data Matrix

full(occur)

5×5 Array{Float64,2}:
 1.0  1.0  0.0  1.0  0.0
 0.0  0.0  1.0  0.0  0.0
 0.0  1.0  0.0  0.0  0.0
 0.0  0.0  0.0  0.0  1.0
 1.0  0.0  0.0  0.0  0.0

SQLite backend

Connecting to SQLite database that was created in pubmed_search_and_save example

db_path = "$(results_dir)/pubmed_obesity_2010_2012.db";
db_sqlite = SQLite.DB(db_path);

Getting the descriptor to index dictionary and occurence matrix

```@example 3_umls_semantic_occurrences @time labels2ind, occur = umls_semantic_occurrences(db_sqlite, umls_concept);

Descriptor to Index Dictionary


```julia
labels2ind

Dict{String,Int64} with 5 entries:
  "Obesity"                  => 1
  "Pediatric Obesity"        => 2
  "Sleep Apnea, Obstructive" => 3
  "Metabolic Syndrome"       => 4
  "Fibromyalgia"             => 5

Output Data Matrix

full(occur)

5×5 Array{Float64,2}:
 1.0  1.0  0.0  1.0  0.0
 0.0  0.0  1.0  0.0  0.0
 0.0  1.0  0.0  0.0  0.0
 0.0  0.0  0.0  0.0  1.0
 1.0  0.0  0.0  0.0  0.0

This page was generated using Literate.jl.