Skip to main content
Fig. 2 | BMC Bioinformatics

Fig. 2

From: DeepECA: an end-to-end learning framework for protein contact prediction from a multiple sequence alignment

Fig. 2

a One example of weight distribution in the sequences of one MSA for T0843 on the CASP11 dataset. b Accuracy improvement depends on the number of sequences in an MSA. We divided 160 protein domains into five bins according to their lengths. The numbers of proteins in the bins are equal (i.e., 32 protein domains in each bin). c Baseline Model top L accuracy shown against the Weighted MSA Model when we have over 200 homologous sequences and d with fewer than 200 homologous sequences

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us