Epilepsy Comorbidities (using Brown MySQL server)

This script run a PubMed-Comorbidities pipeline using the following characteristics:

Main MeSH Heading: "Epilepsy"
UMLS filtering concepts: "Disease or Syndrome", "Mental or Behavioral Dysfunction" or "Neoplastic Process"
Articles analysed: All MEDLINE 2017AA articles tagged with the as a MeSH Heading. Note that this is equivalent to searching PubMed using [MH:noexp] Total number of articles found: 66720
UMLS concept filtering: Comorbidities are analysed on all other MeSH descriptors associated with the specified UMLS concept
This script uses Brown MySQL databases:
- medline
- umls_meta
- pubmed_comorbidities



In [10]:

    
#Settings
const mh = "Anemia"
const concepts = ("Disease or Syndrome", "Mental or Behavioral Dysfunction", "Neoplastic Process");









    



WARNING: redefining constant mh
WARNING: redefining constant concepts

Retrieve Results and Analyze Simple Occurrences



In [11]:

    
using Revise #used during development to detect changes in module
using PubMedMiner

@time occurrence_df = get_semantic_occurrences_df(mh, concepts...);









    



123.160312 seconds (3.71 M allocations: 73.901 MiB, 0.04% gc time)



In [12]:

    
@time stats = mesh_stats(occurrence_df, 20);









    



  1.737753 seconds (1.56 M allocations: 1.345 GiB, 46.08% gc time)



In [13]:

    
PubMedMiner.plot_bar_topn(stats.topn_mesh_labels, stats.topn_mesh_counts)









    Out[13]:

Analyze and Plot Pair Statistics

Correlation Coefficient
Mutual information
Co-occurrence matrix

Correlation Coefficient



In [14]:

    
PubMedMiner.plot_stat_mat(stats.corrcoef, stats.topn_mesh_labels)









    Out[14]:

Pointwise Mutual Information



In [15]:

    
PubMedMiner.plot_stat_mat(stats.pmi_sp, stats.topn_mesh_labels)









    Out[15]:

Co-Occurrences Matrix



In [16]:

    
PubMedMiner.plot_chord_coo(stats.top_coo_sp, stats.topn_mesh_labels)









    Out[16]:

Frequent Item Sets

Visualization of Frequent Item Sets

Basic visualization of frequent item sets using Sankey diagram (experimental - use with caution)
Future work includes better layout for more links as well as the ability to dinamically change the number of of itemsets



In [17]:

    
PubMedMiner.plot_sankey_arules(stats.sankey_sources, stats.sankey_targets, stats.sankey_vals, stats.mesh_names)









    Out[17]: