Table 1.

Results of the test with the first dataset to gauge the accuracy of the NNPIF. For each target, we show its PDB (Protein Data Bank) ID and length. We show Global Distance Test (GDT) scores for the best model, the average GDT over all of the models, the GDT for the best of the five final models selected via clustering and the RMSD for the best model and the average of the mutual GDT scores between the 1,000 reconstructed models. Avg., average; Clust., clustering.

Id.	Length		GDT Max	GDT Avg.	GDT Clust.	RMSDbest	m
2E5TA	46	0.4457	0.3651	0.4022	17.10	0.4942
2EKFA	61	0.3975	0.3073	0.3975	10.17	0.3782
2K0NA	85	0.3824	0.2709	0.3	6.83	0.2891
2PQRC	40	0.5125	0.3792	0.4125	4.91	0.4701
2QFAB	62	0.5202	0.3853	0.4556	5.58	0.4269
2QKHA	94	0.2926	0.2084	0.2394	9.92	0.2458
2QZGA	88	0.3239	0.228	0.2528	14.83	0.2742
2VC8A	72	0.3854	0.2689	0.3021	7.13	0.2801
2YVRA	45	0.4722	0.3613	0.4	5.40	0.4267
3BRTB	61	0.4016	0.3071	0.3607	18.87	0.3921
1BDSA	45	0.5233	0.3776	0.4709	4.01	0.4031
1BH8A	47	0.5944	0.433	0.5111	4.61	0.4338
2EQFA	48	0.4674	0.3322	0.3804	5.22	0.3815
2JEEA	80	0.4423	0.3116	0.3686	6.64	0.368
2JP7A	59	0.5088	0.3369	0.4211	5.36	0.3688
2P9XA	100	0.3316	0.2186	0.2423	10.04	0.2353
2R9IA	73	0.5486	0.4071	0.4514	10.98	0.4336
2V6YA	77	0.4433	0.3009	0.3367	7.31	0.3132
2YQPA	62	0.4583	0.3311	0.3958	10.42	0.3505
2YRMA	45	0.4593	0.3399	0.4128	8.94	0.3618
3CJHA	63	0.5962	0.3908	0.4279	4.53	0.4022