Identification and Development of Synovial B-Cell-Related Genes Diagnostic Signature for Rheumatoid Arthritis

Figure 4

Development of BRGs diagnostic signature by machine learning. (a and c) LASSO coefficient profiles are determined by the optimal lambda. (b and d) Selection of the tuning parameter (lambda) in the LASSO model by the area under the curve (AUC). (e–h) Receiver-operating characteristic curves for nine machine learning models in different comparison groups. (i and j) Forest plot of the AUC score of the nine models in different comparison groups.

(a)

(b)

(c)

(d)

(e)

(f)

(g)

(h)

(i)

(j)