Application of Different Biostatistical Methods in Biological Data Analysis | ||||
| Zagazig Veterinary Journal | ||||
| Article 8, Volume 47, Issue 2, June 2019, Page 203-212 PDF (712.15 K) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/zvjz.2019.11121.1034 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Khairy El-Bayomi; Fatma Mohamed; Mahmoud Eltarabany; Hagar Gouda 
| ||||
| Animal Wealth Development Department, Faculty of Veterinary Medicine, Zagazig University, 44511, Egypt | ||||
| Abstract | ||||
| Logistic regression is one of the popular methods used in genetic data analysis. That is applied to predict a categorical binary dependent variable on basis of predictor variables, and to test the probability of getting a particular value of the dependent variable that is related to the explanatory variable. The objective of this study is to highlight the crucial role of biostatistical methods in increasing the accuracy of the results in veterinary and biological practices. Statistical analysis of previously published data in the National Research Center, Dokki, Giza, Cairo, Egypt was done using SPSS version, 24 to predict hepatocellular carcinoma metastasis by knowing the genotypes, age, and gender of the patients. The genotypes and gender displayed a significant effect on metastasis (P < 0.05) while age had no significant effect on metastasis (P > 0.05). There are other types of data (animal breeding and production) which were analyzed by repeated measures ANOVA and principal component analysis (PCA). The repeated measures ANOVA is equivalent to normalized ANOVA, but for related, not independent groups. Data of this test was obtained from a study aimed to measure body weight of three breeds of rabbits at 4 time points 4th, 6th, 8th and 10th weeks of the experiment. The main effect of breed types of rabbits was significant (P < 0.05), the time (weeks) was highly significant (P < 0.001) and their interaction was also highly significant (P < 0.001). Principal component analysis (PCA) is used to reduce a large set of variables to a small set that still contains most of the information in the large set. A reduced set is easier to analyze and interpret. Data with 6 variables reduced to only 2 variables where initial eigenvalues were > 1 for two variables and their values were (2.768 and 1.147). | ||||
| Keywords | ||||
| principal component analysis (PCA); Repeated measures ANOVA; Sphericity assumption; Logistic regression; Odds ratio | ||||
|
Statistics Article View: 360 PDF Download: 362 |
||||
