Table 3.
Taxa differed between the lowest quartile group (Q1) and the highest quartile group (Q4) for the skeletal muscle mass index in males.
| Taxa levels | Taxonomic assignment a | ANCOM‐II | MaAsLin2 | |
|---|---|---|---|---|
| W b | Coefficient c | P value | ||
| Phylum | — | — | — | — |
| Class | — | — | — | — |
| Order | p_Proteobacteria;c_Gammaproteobacteria;o_Pasteurellales | 25 | 1.908 | 2.1 × 10 −2 |
| Family | p_Firmicutes;c_Negativicutes;o_Acidaminococcales;f_Acidaminococcaceae | 56 | 0.904 | 7.9 × 10−1 |
| p_Firmicutes;c_Clostridia;o_Eubacteriales;_ | 49 | 2.543 | 8.1 × 10 −3 | |
| p_Proteobacteria;c_Gammaproteobacteria;o_Pasteurellales;f_Pasteurellaceae | 41 | 1.908 | 2.1 × 10 −2 | |
| Genus | p_Firmicutes;c_Clostridia;o_Eubacteriales;_;_ | 157 | 2.546 | 8.1 × 10 −3 |
| p_Proteobacteria;c_Gammaproteobacteria;o_Pasteurellales;f_Pasteurellaceae;g_Haemophilus | 146 | 1.909 | 2.1 × 10 −2 | |
| p_Firmicutes;c_Clostridia;o_Eubacteriales;f_Lachnospiraceae;_ | 132 | 1.641 | 3.1 × 10 −2 | |
| Species | p_Firmicutes;c_Clostridia;o_Eubacteriales;_;_;_ | 289 | 2.541 | 8.3 × 10 −3 |
| p_Proteobacteria;c_Gammaproteobacteria;o_Pasteurellales;f_Pasteurellaceae;g_Haemophilus;s_parainfluenzae | 258 | 1.910 | 2.0 × 10 −2 | |
| p_Firmicutes;c_Clostridia;o_Eubacteriales;f_Lachnospiraceae;_;_ | 240 | 1.627 | 3.3 × 10 −2 | |
| p_Firmicutes;c_Clostridia;o_Eubacteriales;f_Lachnospiraceae;g_Roseburia;s_faecis | 218 | 1.536 | 3.2 × 10 −2 | |
| p_Firmicutes;c_Clostridia;o_Eubacteriales;f_Lachnospiraceae;g_Blautia;s_faecis | 201 | 1.132 | 5.4 × 10−1 | |
Note: Both ANCOM‐II and MaAsLin2 models were adjusted for age, BMI and regular physical activity. The Q4 group was set as the reference group and compared with the Q1 group. If the significance in the ANCOM‐II model is validated in the MaAsLin2 model, the taxa and P values are shown in bold.
Abbreviations: ANCOM, analysis of composition of microbiomes; BMI, body mass index; c_, class; f_, family; FDR, false discovery rate; g_, genus; MaAsLin2, multivariate association with linear models; NCBI‐RefSeq, National Center for Biotechnology Information (NCBI) Nucleotide and Taxonomy databases; o_, order; p_, phylum; s_, species.
NCBI‐RefSeq database was used for taxonomic assignment. No. of phylum: 13, no. of class: 26, no. of order: 35, no. of family: 67, no. of genera: 186 and no. of species: 325 in Q1 (n = 155) versus Q4 (n = 156) comparison analysis.
W = X for taxon k, and then H 0k is rejected X times. The W statistic for the significantly different taxa relative to more than 60% other taxa in each taxa level is represented (FDR P < 0.05, Benjamini–Hochberg method).
Exponentiation of the coefficients for the log‐transformed relative abundance of each taxon in the linear model adjusted for age, BMI and regular physical activity using MaAsLin2.