. Author manuscript; available in PMC: 2017 Sep 1.

Published in final edited form as: Proteins. 2015 Sep 18;84(Suppl 1):233–246. doi: 10.1002/prot.24918

Table 5.

Comparison of the first model generated by different pipelines.

Pipeline	TM-score^a	GDT-TS^a	GDT-HA^a	TM_template^b
LOMETS	0.613	0.532	0.387
QUARK-TBM	0.634	0.546	0.393	0.798 (0.644)
I-TASSER^c	0.642 (0.011)	0.553 (0.016)	0.394 (0.396)	0.763 (0.637)
I-TASSER+QUARK^d	0.650 (0.022)	0.561 (0.021)	0.401 (0.041)	0.773 (0.639)

TM-, GDT-TS and GDT-HA scores of the first models compared to the native. The values in parenthesis are the p-value compared to the models generated by the previous row (i.e., p-value of I-TASSER vs.QUARK -TBM, and I-TASSER-QUARK vs. I-TASSER).

TM-score of the first model to the closest template. The value in parenthesis is the average TM-score of the first model to the top 30 closest templates.

I-TASSER without using QUARK-TBM models

I-TASSER predictions using QUARK-TBM models as input. These models were finally submitted to CASP.