A graph theory approach to analyze birth defect associations

ELIAS, DARIO; CAMPAÑA, HEBE; POLETTA, FERNANDO; HEISECKE, SILVINA; GILI, JUAN; RATOWIECKI, JULIA; GIMENEZ, LUCAS; PAWLUK, MARIELA; SANTOS, MARIA RITA; COSENTINO, VIVIANA; URANGA, ROCIO; RITTLER, MONICA; LOPEZ CAMELO, JORGE

PLOS ONE

PUBLIC LIBRARY SCIENCE

Año: 2020 vol. 15

1932-6203

Birth defects are prenatal morphological or functional anomalies.Associations among them are studied to identify their etiopathogenesis. Thegraph theory methods allow analyzing relationships among a complete set ofanomalies. A graph consists of nodes which represent the entities (birthdefects in the present work), and edges that join nodes indicating therelationships among them. The aim of the present study was to validate thegraph theory methods to study birth defect associations. All birth defectsmonitoring records from the Estudio Colaborativo Latino Americano deMalformaciones Congénitas gathered between 1967 and 2017 were used. From around5 million live and stillborn infants, 170,430 had one or more birth defects.Volume-adjusted Chi-Square was used to determine the association strength betweentwo birth defects and to weight the graph edges. The complete birth defectgraph showed a Log-Normal degree distribution and its characteristics differedfrom random, scalefree and small-world graphs. The graph comprised 118 nodesand 550 edges. Birth defects with the highest centrality values werenonspecific codes such as Other upper limb anomalies. After partition, thegraph yielded 12 groups; most of them were recognizable and included conditionssuch as VATER and OEIS associations, and Patau syndrome. Our findings validatethe graph theory methods to study birth defect associations. This method may contributeto identify underlying etiopathogeneses as well as to improve coding systems.