A novel convolution attention model for predicting transcription factor binding sites by combination of sequence and shape.

The discovery of putative transcription factor binding sites (TFBSs) is important for understanding the underlying binding mechanism and cellular functions. Recently, many computational methods have been proposed to jointly account for DNA sequence and shape properties in TFBSs prediction. However, these methods fail to fully utilize the latent features derived from both sequence and shape profiles and have limitation in interpretability and knowledge discovery. To this end, we present a novel Deep Convolution Attention network combining Sequence and Shape, dubbed as D-SSCA, for precisely predicting putative TFBSs. Experiments conducted on 165 ENCODE ChIP-seq datasets reveal that D-SSCA significantly outperforms several state-of-the-art methods in predicting TFBSs, and justify the utility of channel attention module for feature refinements. Besides, the thorough analysis about the contribution of five shapes to TFBSs prediction demonstrates that shape features can improve the predictive power for transcription factors-DNA binding. Furthermore, D-SSCA can realize the cross-cell line prediction of TFBSs, indicating the occupancy of common interplay patterns concerning both sequence and shape across various cell lines. The source code of D-SSCA can be found at https://github.com/MoonLord0525/.

Genetic variability, phylogeny and functional implication of the long control region in human papillomavirus type 16, 18 and 58 in Chengdu, China.

BACKGROUND: Long control region (LCR) of human papillomavirus (HPV) has shown multiple functions on regulating viral transcription. The variations of LCR related to different lineages/sub-lineages have been found to affect viral persistence and cervical cancer progression differently. In this study, we focused on gene polymorphism of HPV16/18/58 LCR to assess the effect variations caused on transcription factor binding sites (TFBS) and provided more data for further study of LCR in Southwest China. METHODS: LCR of HPV16/18/58 were amplified and sequenced to do polymorphic and phylogenetic anlysis. Sequences of each type were aligned with the reference sequence by MEGA 6.0 to identify SNPs. Neighbor-joining phylogenetic trees were constructed using MEGA 6.0. Transcription factor binding sites were predicted by JASPAR database. RESULTS: The prevalence of these three HPVs ranked as HPV16 (12.8%) > HPV58 (12.6%) > HPV18 (3.5%) in Chengdu, Southwest China. 59 SNPs were identified in HPV16-LCR, 18 of them were novel mutations. 30 SNP were found in HPV18-LCR, 8 of them were novel. 55 SNPs were detected in HPV58-LCR, 18 of them were novel. Also, an insertion (CTTGTCAGTTTC) was detected in HPV58-LCR between position 7279 and 7280. As shown in the neighbor-joining phylogenetic trees, most isolates of HPV16/18/58 were clustered into lineage A. In addition, one isolate of HPV16 was classified into lineage C and 3 isolates of HPV58 were classified as lineage B. JASPAR results suggested that TFBS were potentially influenced by 7/6 mutations on LCR of HPV16/18. The insertion and 5 mutations were shown effects in LCR of HPV58. CONCLUSION: This study provides more data for understanding the relation among LCR mutations, lineages and carcinogenesis. It also helps performing further study to demonstrate biological function of LCR and find potential marker for diagnosis and therapy.

Genetic diversity and functional implication of the long control region in human papillomavirus types 52, 58, and 16 from Central China.

OBJECT: High-risk human papillomavirus (HR-HPV) is a main reason for cervical cancer. The long control region (LCR) of the genome plays a variety of roles in the transcription of the virus. METHODS: LCR sequences were amplified by polymerase chain reaction (PCR) and confirmed by DNA sequencing. MEGA 11.0 software and NCBI blast were used to analyze the sequences and construct the Neighbor-Joining tree. In addition, the JASPAR database was used to predict the potential transcription factor binding sites (TFBS). RESULTS: For HPV-52 LCR, 68 single nucleotide polymorphisms (SNPs), 8 deletions, and 1 insertion were found, 17 of which were novel variations. Most of the variants were clustered in B2 sub-lineage (96.22%). For HPV-58 LCR, 25.43% of samples were prototype. 49 SNPs, 2 deletions, and 1 insertion were observed in the remaining samples. A1 sub-lineage was the most frequent (64.16%). For HPV-16 LCR, 75 SNPs and 2 deletions were identified, 13 of which were newly identified. A total of 55.68% of the variants were distributed in A4 sub-lineage. The JASPAR results suggested that multiple variations occurred in TFBSs, which might affect the function of transcription factors. CONCLUSIONS: This study provides experimental data for further studies on the epidemiology and biological function of LCR. Various LCR mutational data may prove useful for exploring the carcinogenic mechanism of HPV.