The article introduces ANCOM-BC, a method for analyzing microbiome data with bias correction. It addresses the challenge of differential abundance (DA) analysis in microbiome data, which is complicated by differences in sampling fractions across samples. ANCOM-BC estimates sampling fractions and corrects bias, using a linear regression framework to model absolute abundance data. This approach provides statistically valid tests, confidence intervals, and controls the False Discovery Rate (FDR), while maintaining adequate power and computational efficiency. Existing methods like ANCOM and DR have limitations, such as not providing p-values or confidence intervals. ANCOM-BC improves upon these by incorporating sampling fraction correction, allowing for accurate DA analysis. It is particularly effective in scenarios with varying sampling fractions, as demonstrated through simulations and real data analysis. The method was tested on synthetic data and real global gut microbiota data, showing superior performance in controlling FDR and maintaining power compared to other methods. ANCOM-BC also provides confidence intervals for differential abundance, which is crucial for interpreting results. The study highlights the importance of accounting for sampling fractions in microbiome data analysis, as differences in sampling fractions can introduce bias and affect the accuracy of DA results. ANCOM-BC addresses this by incorporating a sample-specific offset term in a linear regression framework, which adjusts for sampling fraction differences. The method was applied to gut microbiota data from the USA, Malawi, and Venezuela, revealing significant differences in the abundance of certain taxa between populations. These findings underscore the importance of accurate DA analysis in understanding microbial community composition and its relationship to environmental and demographic factors. In conclusion, ANCOM-BC offers a robust solution for DA analysis in microbiome data by addressing the critical issue of sampling fraction bias. It provides statistically valid results, maintains high power, and is computationally efficient, making it a valuable tool for researchers in the field of microbiome studies.The article introduces ANCOM-BC, a method for analyzing microbiome data with bias correction. It addresses the challenge of differential abundance (DA) analysis in microbiome data, which is complicated by differences in sampling fractions across samples. ANCOM-BC estimates sampling fractions and corrects bias, using a linear regression framework to model absolute abundance data. This approach provides statistically valid tests, confidence intervals, and controls the False Discovery Rate (FDR), while maintaining adequate power and computational efficiency. Existing methods like ANCOM and DR have limitations, such as not providing p-values or confidence intervals. ANCOM-BC improves upon these by incorporating sampling fraction correction, allowing for accurate DA analysis. It is particularly effective in scenarios with varying sampling fractions, as demonstrated through simulations and real data analysis. The method was tested on synthetic data and real global gut microbiota data, showing superior performance in controlling FDR and maintaining power compared to other methods. ANCOM-BC also provides confidence intervals for differential abundance, which is crucial for interpreting results. The study highlights the importance of accounting for sampling fractions in microbiome data analysis, as differences in sampling fractions can introduce bias and affect the accuracy of DA results. ANCOM-BC addresses this by incorporating a sample-specific offset term in a linear regression framework, which adjusts for sampling fraction differences. The method was applied to gut microbiota data from the USA, Malawi, and Venezuela, revealing significant differences in the abundance of certain taxa between populations. These findings underscore the importance of accurate DA analysis in understanding microbial community composition and its relationship to environmental and demographic factors. In conclusion, ANCOM-BC offers a robust solution for DA analysis in microbiome data by addressing the critical issue of sampling fraction bias. It provides statistically valid results, maintains high power, and is computationally efficient, making it a valuable tool for researchers in the field of microbiome studies.