Table 1.
Baseline characteristics of RA and non-RA groups.
Figure 1.
OPLS-DA of the metabolite profiles of RA and non-RA groups.
(a) Score plot of the OPLS-DA model for RA and non-RA groups (t[1]P, score of the non-orthogonal component; t[2]O, score of the orthogonal component). (b) V-plot with p(corr) and VIP values of 105 metabolites. The metabolites with p(corr) <0 were those decreased in RA groups while the metabolites with p(corr) >0 were those increased in RA groups.
Figure 2.
HCA of 105 metabolites from synovial fluid samples of RA and non-RA patients.
Each column and row represents a disease and an individual metabolite, respectively.
Figure 3.
ROC analysis of the predictive power of the 20 combined biomarkers for distinguishing RA and non-RA groups.
A sensitivity and specificity were 92.3% and 68.0%, respectively, and the value of AUC was 0.812.
Table 2.
VIP and AUC values of the metabolites that significantly contribute to the discrimination between the RA and non-RA groups.
Figure 4.
Schematic comparison of the primary metabolisms of RA vs. non-RA groups (AS, BD, and GO).
The box and whisker plots indicate the intracellular metabolite levels for each disease group (red, increased in RA; green, increased in non-RA). AcCOA, acetyl-CoA; ALA, alanine; ARG, arginine; ARG-SUC, arginine-succinate; ASN, asparagine; ASP, aspartate; CIT, citrate; CITR, citrulline; CMP, carbamoyl phosphate; FAs, fatty acids; FUM, fumarate; GLC, glucose; GLN, glutamine; GLU, glutamate; αKG, α-ketoglutarate; LYS, lysine; MAL, malate; OA, oxalate; ORNT, ornithine; PEP, phosphoenolpyruvate; PHA, phenylalanine; PRO, proline; SER, serine; SUCC, succinate; TRP, tryptophan; TYR, tyrosine.