The article introduces a methodology called Analysis of Compositions of Microbiomes with Bias Correction (ANCOM-BC) to address the challenges in differential abundance (DA) analysis of microbiome data. The main issue is the bias introduced by differences in sampling fractions across samples. ANCOM-BC estimates these unknown sampling fractions and corrects the bias using a linear regression framework, which models absolute abundance data. This approach provides several advantages over existing methods, including: 1. **Statistical Validity**: It provides statistically valid tests with appropriate p-values. 2. **Confidence Intervals**: It provides confidence intervals for differential abundance of each taxon. 3. **False Discovery Rate (FDR) Control**: It controls the FDR while maintaining adequate power. 4. **Computational Simplicity**: It is computationally simple to implement. The article also discusses the limitations of existing methods, such as ANOVA, Kruskal–Wallis test, metagenomeSeq, ANCOM, and Differential Ranking (DR), which often suffer from inflated FDR or lack of statistical validity. ANCOM-BC is evaluated through simulation studies and real data analysis, demonstrating its effectiveness in controlling FDR and maintaining high power. The method is particularly useful for analyzing microbiome data with compositional characteristics, where standard statistical methods are not appropriate.The article introduces a methodology called Analysis of Compositions of Microbiomes with Bias Correction (ANCOM-BC) to address the challenges in differential abundance (DA) analysis of microbiome data. The main issue is the bias introduced by differences in sampling fractions across samples. ANCOM-BC estimates these unknown sampling fractions and corrects the bias using a linear regression framework, which models absolute abundance data. This approach provides several advantages over existing methods, including: 1. **Statistical Validity**: It provides statistically valid tests with appropriate p-values. 2. **Confidence Intervals**: It provides confidence intervals for differential abundance of each taxon. 3. **False Discovery Rate (FDR) Control**: It controls the FDR while maintaining adequate power. 4. **Computational Simplicity**: It is computationally simple to implement. The article also discusses the limitations of existing methods, such as ANOVA, Kruskal–Wallis test, metagenomeSeq, ANCOM, and Differential Ranking (DR), which often suffer from inflated FDR or lack of statistical validity. ANCOM-BC is evaluated through simulation studies and real data analysis, demonstrating its effectiveness in controlling FDR and maintaining high power. The method is particularly useful for analyzing microbiome data with compositional characteristics, where standard statistical methods are not appropriate.