Skip to main content

Table 8 Performance of DN ensembles composed of varying architectures on CASP9 targets

From: A study and benchmark of DNcon: a method for protein residue-residue contact prediction using deep networks

 

Acc. Top L/5 (SE)

Acc. Top L (SE)

Architecture

Long

Medium

Long

Medium

500-500-500-350-1

0.174(.012)

0.245(.014)

0.113(.006)

0.144(.008)

750-500-350-1

0.162(.012)

0.231(.013)

0.101(.006)

0.137(.007)

500-500-350-1

0.182(.012)

0.24(.013)

0.122(.006)

0.150(.007)

500-250-1

0.159(.012)

0.243(.131)

0.107(.006)

0.142(.007)

250-250-1

0.169(.010)

0.236(.012)

0.108(.006)

0.142(.007)