Fig. 4From: MalariaSED: a deep learning framework to decipher the regulatory contributions of noncoding variants in malaria parasitesTF binding effects of genomic variants predicted by the DL framework demonstrate the contribution of previously identified sequence motifs. A Potential single-nucleotide substitutions (SNSs) at reported pfAP2-G motif, GTRC (committed schizonts), and GTAC (sexual rings and stage I gametocytes) significantly reduce the binding affinity of PfAP2-G. Comparisons are between reported PfAP2-G motifs and all 4-mers (ALL) appeared on predicted ApAP2-G binding sites. B The PfAP2-I binding effects of base substitutions at the reported motif GTGCA are significant compared with all 5-mers (ALL) in P. falciparum genome during schizonts. The other two motifs (GGTCG and CTTGC) from PfAP2-I binding domains 1 and 2 are not able to show significantly reduced PfAP2-I affinity, which is consistent with experimental results [5]. C The significantly reduced binding affinity of SNSs at reported PfBDP1 binding motif GTGCA in schizonts supports the previous conclusion that PfAP2-I and PfBDP1 may form a protein complex to bind DNA. However, SNSs at the GTGCA motif are not likely to disrupt PfBDP1 binding at the trophozoite stage. D, E Single-nucleotide substitutions at two reported binding motifs in P. Berghei significantly weaken PbAP2-G2 and PbAP2-O binding sites. ***Wilcoxon test compared with whole-genome background p < 2.2e − 16Back to article page