In this project, Zheng et al. use the biweight midcorrelation for constructing binary networks. They employ Biweight Midcorrelation, a measure less sensitive to outliers than the traditional Pearson correlation, to assess the similarity between gene expression profiles. The first paper used the method called the ‘half-thresholding’ strategy1, and another study utilised the ‘Differential Coexpression Threshold’ strategy2 to eliminate non-informative correlation pairs, thus enabling the calculation of the differential coexpression value for each gene.
Below is the code based on the formulas from the papers
1. Biweight midcorrelation coefficients Function
2. Functions for the 'half-thresholding'
3. Functions for the Maximum Clique Concept & k-Clique Algorithm
[5BMKC(Pending)]
The formula from the biweight midcorrelation and half-thresholding (BMHT) algorithm was obtained from the paper1 through the Attribution 4.0 International (CC BY 4.0) license, and the biweight midcorrelation and k-clique (BMKC) algorithm was retrieved from the paper2 via the Attribution 3.0 International (CC BY 3.0). These licenses are based on the corresponding journals' copyright statements.
Ebru Temizhan, Hamit Mirtagioglu, & Mehmet Mendes. (2022). Which Correlation Coefficient Should Be Used for Investigating Relations between Quantitative Variables?. American Scientific Research Journal for Engineering, Technology, and Sciences, 85(1), 265–277. Retrieved from https://asrjetsjournal.org/index.php/American_Scientific_Journal/article/view/7326
Song, L., Langfelder, P., & Horvath, S. (2012). Comparison of co-expression measures: mutual information, correlation, and model based indices. BMC bioinformatics, 13, 328. https://doi.org/10.1186/1471-2105-13-328
Despite the wealth of models and formulas presented in scientific publications, there is often a lack of progression into programming packages that can be utilised by others. "BioTranslate: Turning Bioscience Publications into Applications" is a project dedicated to bridging the gap between theoretical knowledge and practical application in the field of biosciences.
Footnotes
-
Zheng, C. H., Yuan, L., Sha, W., & Sun, Z. L. (2014). Gene differential coexpression analysis based on biweight correlation and maximum clique. BMC bioinformatics, 15 Suppl 15(Suppl 15), S3. https://doi.org/10.1186/1471-2105-15-S15-S3 ↩ ↩2
-
Yuan, L., Zheng, C. H., Xia, J. F., & Huang, D. S. (2015). Module Based Differential Coexpression Analysis Method for Type 2 Diabetes. BioMed research international, 2015, 836929. https://doi.org/10.1155/2015/836929 ↩ ↩2