Table I
Clinical characteristics of patients and healthy controls (average ± standard deviation).
| GD (n = 27) | HT (n = 27) | Controls (n = 16) | |
|---|---|---|---|
| Age (years) | 49.20 ± 8.68 | 56.77 ± 12.44 | 49.31 ± 13.36 |
| Sex (M/F) | 8/19 | 11/16 | 7/9 |
| FT3 (pmol/l) | 14.74 ± 8.65** | 3.93 ± 1.22 | 5.13 ± 0.76 |
| FT4 (pmol/l) | 52.19 ± 24.83** | 7.73 ± 2.99* | 17.91 ± 1.88 |
| TSH (mIU/l) | 0.005 ± 0.000** | 38.798 ± 32.452** | 3.030 ± 0.806 |
| ATG (IU/ml) | 371.84 ± 320.30** | 1248.39 ± 2623.73** | 56.72 ± 26.04 |
| ATPO (IU/ml) | 352.04 ± 148.07** | 519.40 ± 833.86** | 12.27 ± 8.43 |
| TRAb (IU/ml) | 8.69 ± 2.90** | 1.21 ± 0.66 | 0.68 ± 0.2 |
Fig. 1
The gut microbiota of GD and HT patients were different from that of the healthy control group.
A) The rank-abundance curve of the GD group, B) the rank-abundance curve of the HT group.
Fig. 1
The gut microbiota of GD and HT patients were different from that of the healthy control group.
C) histogram of horizontal flora composition of “family”, D) histogram of horizontal flora composition of “genus”, E) PlS-DA analysis with group supervision.
Fig. 1
The gut microbiota of GD and HT patients were different from that of the healthy control group.
F) ANOSIM analysis.
Fig. 2
Bacterial flora classification map obtained by LEfSe analysis.
A) LEfSe shows the greatest difference in abundance (taxa) between the three groups (LDA threshold > 3).
Fig. 2
Bacterial flora classification map obtained by LEfSe analysis.
B–G) the difference in microbiota between the GD group or HT groups and the healthy control group at the phylum level (B, C), at the family level (D, E), and at the genus level (F, G). *p < 0.05; ** p < 0.01; ***p < 0.001.
Fig. 2
E, F, G
Fig. 3
Random forest analysis and validation information.
A) Random forest analysis between the GD and healthy control groups, and B) between the HT group and control groups.
Fig. 3
Random forest analysis and validation information.
C) verification information of the first three genera of random forest results from the GD group and healthy control group, and D) between the HT group and healthy control group.
Fig. 4
Prediction Results using the COG and KEGG databases.
A, B) The difference in the COG functional prediction between the disease and control groups; C, D) the difference in the KEGG function prediction between the disease and control groups; E, F) the difference in the COG abundance prediction between the disease and control groups; G, H) the difference in the KEGG enzyme prediction between the disease and the control groups. *p < 0.05; **p < 0.01, ***p < 0.001.
ko02010 – ABC transporters, ko00230 – purine metabolism, ko00520 – amino sugar and nucleotide sugar metabolism, ko02020 – two-component system, ko00330 – arginine and proline metabolism, ko00970 – aminoacyl-tRNA biosynthesis, ko00500 – starch and sucrose metabolism, ko00680 – methane metabolism, ko00250 – alanine, aspartate and glutamate metabolism, ko00010 – glycolysis/gluconeogenesis, ko00190 – oxidative phosphorylation, ko00860 – porphyrin and chlorophyll metabolism, ko00270 – cysteine and methionine metabolism, ko00720 – carbon fixation pathways in prokaryotes, ko00620 – pyruvate metabolism, ko03010 – ribosome, ko00240 – pyrimidine metabolism, ko03440 – homologous recombination.
Fig. 4
Prediction Results using the COG and KEGG databases.
C, D) the difference in the KEGG function prediction between the disease and control groups. *p < 0.05; **p < 0.01, ***p < 0.001.
Fig. 4
Prediction Results using the COG and KEGG databases.
E, F) the difference in the COG abundance prediction between the disease and control groups.
*p < 0.05; **p < 0.01, ***p < 0.001.
ko02010 – ABC transporters, ko00230 – purine metabolism, ko00520 – amino sugar and nucleotide sugar metabolism, ko02020 – two-component system, ko00330 – arginine and proline metabolism, ko00970 – aminoacyl-tRNA biosynthesis, ko00500 – starch and sucrose metabolism, ko00680 – methane metabolism, ko00250 – alanine, aspartate and glutamate metabolism, ko00010 – glycolysis/gluconeogenesis, ko00190 – oxidative phosphorylation, ko00860– porphyrin and chlorophyll metabolism, ko00270 – cysteine and methionine metabolism, ko00720 – carbon fixation pathways in prokaryotes, ko00620 – pyruvate metabolism, ko03010 – ribosome, ko00240 – pyrimidine metabolism, ko03440 – homologous recombination.
Fig. 4
Prediction Results using the COG and KEGG databases.
G, H) the difference in the KEGG enzyme prediction between the disease and the control groups.
*p < 0.05; **p < 0.01, ***p < 0.001.
Fig. 5
Diagram of random forest differential strains.