Premium
Genotype Discrimination Using Facial and Body Anthropometry among Hausas
Author(s) -
Taura Magaji Garba,
Adamu Lawan Hassan,
Gudaji Abdullahi
Publication year - 2016
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.30.1_supplement.555.2
Subject(s) - anthropometry , logistic regression , receiver operating characteristic , population , statistics , genotype , regression analysis , medicine , demography , mathematics , biology , genetics , sociology , gene , environmental health
Individual's genotype can be classified as normal, carrier or sickle cell patient when it is AA, AS and SS respectively. Several researches attempted to discriminate the sickle cell patient with normal subject using anthropometry, however less attention is paid to the discrimination of an individual with normal genotype from carrier especially among Hausa population. The present study aimed at discriminating between normal and carrier from Hausa ethnic population using body and facial anthropometry. A total of 219 participants with mean age of 20.41 ± 2.94 years were involved. The genotype analyses revealed 197 and 22 participants as normal and carrier respectively. Direct anthropometry was used to measure the height, weight and 17 facial features. Independent sample t – test was used for the test of differences. Binary logistic regression (BLR) analysis was used to obtain a model for genotype discrimination. The predicted probabilities of BLR for the variables that contributed best to the prediction were analyzed using Receiver Operating Characteristic (ROC) curve. The results indicated that none of the facial features showed significant difference between normal and carrier. Of the three parameters of height, weight and BMI only weight exhibited significant differences between the groups. The model prediction accuracy with all variables in the equation was 90.9% with only 19.7% contribution to the prediction (Nagelkerke R Square = 0.197). Using stepwise BLR only weight was included in the equation (GENOTYPE= 0.058×WT −1.001, cut value is 0.5, Wald statistics= 5.022, p = 0.025) with overall accuracy of 90%. The ROC curve analyses for the weight only showed the area under curve (AUC) of 0.654 indicating 65.4% best prediction probability of the model. In conclusion, weight demonstrated a superior potential in discriminating the genotype over other variables considered in the study among Hausas of Nigeria.