Limits...
Construction of phosphorylation interaction networks by text mining of full-length articles using the eFIP system.

Tudor CO, Ross KE, Li G, Vijay-Shanker K, Wu CH, Arighi CN - Database (Oxford) (2015)

Bottom Line: The evaluation of eFIP on full-length articles achieved 92.4% precision, 76.5% recall and 83.7% F-measure on 100 article sections.To demonstrate eFIP for knowledge extraction and discovery, we constructed phosphorylation-dependent interaction networks involving 14-3-3 proteins identified from cancer-related versus diabetes-related articles.Comparison of the phosphorylation interaction network of kinases, phosphoproteins and interactants obtained from eFIP searches, along with enrichment analysis of the protein set, revealed several shared interactions, highlighting common pathways discussed in the context of both diseases.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer and Information Sciences and Center for Bioinformatics and Computational Biology, University of Delaware, Newark, DE, USA Department of Computer and Information Sciences and Center for Bioinformatics and Computational Biology, University of Delaware, Newark, DE, USA oanat@udel.edu.

Show MeSH

Related in: MedlinePlus

Distribution of subsections in the eFIP results (blue), RLIMS-P results (red) and the entire PMC OA collection (green).
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4381107&req=5

bav020-F3: Distribution of subsections in the eFIP results (blue), RLIMS-P results (red) and the entire PMC OA collection (green).

Mentions: In our work, a subsection is considered to be any heading together with the text occurring right afterwards and before the next heading in the article. For example, the ‘Abstract’ subsection is composed of the text between the Abstract heading and the ‘Introduction’ heading. Likewise, the heading ‘GST pull down assay’ and the text between the ‘GST pull down assay’ and ‘Blot overlay assay’ headings in Figure 2 constitute a subsection. To each subsection, we also associate the titles/headings of the parent nodes, as these could hold important information complementing the actual text. Thus, the title of the article itself is always listed together with every subsection, and the title of major sections, such as ‘Results’ is also carried over in the subsections. A type is assigned to each subsection, depending on the title of the subsection’s parent, or based on the type of subsection specified in the original XML file. If no type can be determined, either because no annotation could be found in the XML file, or because the parent section has a title other than the typical Introduction, Results, Discussion and so on, then a type of ‘other’ is assigned to the subsection. The types of subsections and their distribution in the local PMC index are as follows: Abstract (8%), Introduction (4.2%), Background (1.5%), Methods & Materials (24%), Results (16.5%), Discussion (6%), Conclusions (1.6%), figure (21%) and other (17.2%). This can be seen in Figure 3.Figure 3.


Construction of phosphorylation interaction networks by text mining of full-length articles using the eFIP system.

Tudor CO, Ross KE, Li G, Vijay-Shanker K, Wu CH, Arighi CN - Database (Oxford) (2015)

Distribution of subsections in the eFIP results (blue), RLIMS-P results (red) and the entire PMC OA collection (green).
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4381107&req=5

bav020-F3: Distribution of subsections in the eFIP results (blue), RLIMS-P results (red) and the entire PMC OA collection (green).
Mentions: In our work, a subsection is considered to be any heading together with the text occurring right afterwards and before the next heading in the article. For example, the ‘Abstract’ subsection is composed of the text between the Abstract heading and the ‘Introduction’ heading. Likewise, the heading ‘GST pull down assay’ and the text between the ‘GST pull down assay’ and ‘Blot overlay assay’ headings in Figure 2 constitute a subsection. To each subsection, we also associate the titles/headings of the parent nodes, as these could hold important information complementing the actual text. Thus, the title of the article itself is always listed together with every subsection, and the title of major sections, such as ‘Results’ is also carried over in the subsections. A type is assigned to each subsection, depending on the title of the subsection’s parent, or based on the type of subsection specified in the original XML file. If no type can be determined, either because no annotation could be found in the XML file, or because the parent section has a title other than the typical Introduction, Results, Discussion and so on, then a type of ‘other’ is assigned to the subsection. The types of subsections and their distribution in the local PMC index are as follows: Abstract (8%), Introduction (4.2%), Background (1.5%), Methods & Materials (24%), Results (16.5%), Discussion (6%), Conclusions (1.6%), figure (21%) and other (17.2%). This can be seen in Figure 3.Figure 3.

Bottom Line: The evaluation of eFIP on full-length articles achieved 92.4% precision, 76.5% recall and 83.7% F-measure on 100 article sections.To demonstrate eFIP for knowledge extraction and discovery, we constructed phosphorylation-dependent interaction networks involving 14-3-3 proteins identified from cancer-related versus diabetes-related articles.Comparison of the phosphorylation interaction network of kinases, phosphoproteins and interactants obtained from eFIP searches, along with enrichment analysis of the protein set, revealed several shared interactions, highlighting common pathways discussed in the context of both diseases.

View Article: PubMed Central - PubMed

Affiliation: Department of Computer and Information Sciences and Center for Bioinformatics and Computational Biology, University of Delaware, Newark, DE, USA Department of Computer and Information Sciences and Center for Bioinformatics and Computational Biology, University of Delaware, Newark, DE, USA oanat@udel.edu.

Show MeSH
Related in: MedlinePlus