Leveraging a disulfidptosis-related signature to predict the prognosis and immunotherapy effectiveness of cutaneous melanoma based on machine learning

Table 1 The quantified importance of prognostic disulfidptosis-related messenger genes by machin learning

	Decision Tree	LASSO	Random Forest	GBDT	XGBoost	AVG
GBP5	0.013827	0.373494	0.215771	0.165928	0.044820	0.162768	(1)
HLA-DQA1	0	0.229167	0.149802	0.112942	0.029389	0.10426	(2)
HLA-DRA	0.069321	0	0.217371	0.202672	0.021325	0.102138	(3)
CD79A	0.043209	0.08349	0.15570	0.074456	0.022523	0.075875	(4)
HE5	0.054514	0.280299	0.023088	0.015352	0.025615	0.070538	(5)
HLA-DMB	0.060051	0.023476	0.115609	0.083555	0.012552	0.059049	(6)
LTB	0.064557	0	0.026341	0.025064	0.027094	0.0286111	(7)
CD2	0.418228	0	0.035964	0.032222	0.025557	0.027113	(8)
GZMA	0.034531	0	0.041947	0.025454	0.030842	0.026555	(9)
CCL5	0.024580	0	0.02469	0.059022	0.012451	0.024148	(10)

ISSN: 1528-3658