A deep learning model to identify gene expression level using cobinding transcription factor signals.

Gene expression is directly controlled by transcription factors (TFs) in a complex combination manner. It remains a challenging task to systematically infer how the cooperative binding of TFs drives gene activity. Here, we quantitatively analyzed the correlation between TFs and surveyed the TF interaction networks associated with gene expression in GM12878 and K562 cell lines. We identified six TF modules associated with gene expression in each cell line. Furthermore, according to the enrichment characteristics of TFs in these TF modules around a target gene, a convolutional neural network model, called TFCNN, was constructed to identify gene expression level. Results showed that the TFCNN model achieved a good prediction performance for gene expression. The average of the area under receiver operating characteristics curve (AUC) can reach up to 0.975 and 0.976, respectively in GM12878 and K562 cell lines. By comparison, we found that the TFCNN model outperformed the prediction models based on SVM and LDA. This is due to the TFCNN model could better extract the combinatorial interaction among TFs. Further analysis indicated that the abundant binding of regulatory TFs dominates expression of target genes, while the cooperative interaction between TFs has a subtle regulatory effects. And gene expression could be regulated by different TF combinations in a nonlinear way. These results are helpful for deciphering the mechanism of TF combination regulating gene expression.

Ribosome footprint profiling enables elucidating the systemic regulation of fatty acid accumulation in Acer truncatum.

BACKGROUND: The accumulation of fatty acids in plants covers a wide range of functions in plant physiology and thereby affects adaptations and characteristics of species. As the famous woody oilseed crop, Acer truncatum accumulates unsaturated fatty acids and could serve as the model to understand the regulation and trait formation in oil-accumulation crops. Here, we performed Ribosome footprint profiling combing with a multi-omics strategy towards vital time points during seed development, and finally constructed systematic profiling from transcription to proteomes. Additionally, we characterized the small open reading frames (ORFs) and revealed that the translational efficiencies of focused genes were highly influenced by their sequence features. RESULTS: The comprehensive multi-omics analysis of lipid metabolism was conducted in A. truncatum. We applied the Ribo-seq and RNA-seq techniques, and the analyses of transcriptional and translational profiles of seeds collected at 85 and 115 DAF were compared. Key members of biosynthesis-related structural genes (LACS, FAD2, FAD3, and KCS) were characterized fully. More meaningfully, the regulators (MYB, ABI, bZIP, and Dof) were identified and revealed to affect lipid biosynthesis via post-translational regulations. The translational features results showed that translation efficiency tended to be lower for the genes with a translated uORF than for the genes with a non-translated uORF. They provide new insights into the global mechanisms underlying the developmental regulation of lipid metabolism. CONCLUSIONS: We performed Ribosome footprint profiling combing with a multi-omics strategy in A. truncatum seed development, which provides an example of the use of Ribosome footprint profiling in deciphering the complex regulation network and will be useful for elucidating the metabolism of A. truncatum seed oil and the regulatory mechanisms.

Transcriptome profiling provides insights into leaf color changes in two Acer palmatum genotypes.

BACKGROUND: Ornamental trees with seasonally-dependent leaf color, such as Acer palmatum, have gained worldwide popularity. Leaf color is a main determinant of the ornamental and economic value of A. palmatum. However, the molecular mechanisms responsible for leaf color changes remain unclear. RESULTS: We chose A. palmatum cultivars with yellow ('Jinling Huangfeng') and red ('Jinling Danfeng') leaves as the ideal material for studying the complex metabolic networks responsible for variations in leaf coloration. The 24 libraries obtained from four different time points in the growth of 'Jinling Huangfeng' and 'Jinling Danfeng' was subjected to Illumina high-throughput sequencing. We observed that the difference in cyanidin and delphinidin content is the primary reason behind the varying coloration of the leaves. Transcriptomic analyses revealed 225,684 unigenes, and the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis of differentially expressed genes (DEGs) confirmed that they were involved in 'anthocyanin biosynthesis.' Eighteen structural genes involved in anthocyanin biosynthesis were thought to be related to anthocyanin accumulation, whereas 46 MYBs, 33 basic helix-loop-helixs (bHLHs), and 29 WD40s were presumed to be involved in regulating anthocyanin biosynthesis. Based on weighted gene co-expression network analysis (WGCNA), three candidate genes (ApRHOMBOID, ApMAPK, and ApUNE10) were screened in the significant association module with a correlation coefficient (r2) of 0.86. CONCLUSION: In this study, the leaf color changes of two A. palmatum genotypes were analyzed. These findings provide novel insights into variations in leaf coloration and suggest pathways for targeted genetic improvements in A. palmatum.

Assembly and comparative analysis of the first complete mitochondrial genome of Acer truncatum Bunge: a woody oil-tree species producing nervonic acid.

BACKGROUND: Acer truncatum (purpleblow maple) is a woody tree species that produces seeds with high levels of valuable fatty acids (especially nervonic acid). The species is admired as a landscape plant with high developmental prospects and scientific research value. The A. truncatum chloroplast genome has recently been reported; however, the mitochondrial genome (mitogenome) is still unexplored. RESULTS: We characterized the A. truncatum mitogenome, which was assembled using reads from PacBio and Illumina sequencing platforms, performed a comparative analysis against different species of Acer. The circular mitogenome of A. truncatum has a length of 791,052 bp, with a base composition of 27.11% A, 27.21% T, 22.79% G, and 22.89% C. The A. truncatum mitogenome contains 62 genes, including 35 protein-coding genes, 23 tRNA genes and 4 rRNA genes. We also examined codon usage, sequence repeats, RNA editing and selective pressure in the A. truncatum mitogenome. To determine the evolutionary and taxonomic status of A. truncatum, we conducted a phylogenetic analysis based on the mitogenomes of A. truncatum and 25 other taxa. In addition, the gene migration from chloroplast and nuclear genomes to the mitogenome were analyzed. Finally, we developed a novel NAD1 intron indel marker for distinguishing several Acer species. CONCLUSIONS: In this study, we assembled and annotated the mitogenome of A. truncatum, a woody oil-tree species producing nervonic acid. The results of our analyses provide comprehensive information on the A. truncatum mitogenome, which would facilitate evolutionary research and molecular barcoding in Acer.

Discovering the key genes and important DNA methylation regions in breast cancer.

BACKGROUND: Breast cancer is the malignant tumor with the highest incidence in women. DNA methylation has an important effect on breast cancer, but the effect of abnormal DNA methylation on gene expression in breast cancer is still unclear. Therefore, it is very important to find therapeutic targets related to DNA methylation. RESULTS: In this work, we calculated the DNA methylation distribution and gene expression level in cancer and para-cancerous tissues for breast cancer samples. We found that DNA methylation in key regions is closely related to gene expression by analyzing the relationship between the distribution characteristics of DNA methylation in different regions and the change of gene expression level. Finally, the 18 key genes (17 tumor suppressor genes and 1 oncogene) related to prognosis were confirmed by the survival analysis of clinical data. Some important DNA methylation regions in these genes that result in breast cancer were found. CONCLUSIONS: We believe that 17 TSGs and 1 oncogene may be breast cancer biomarkers regulated by DNA methylation in key regions. These results will help to explore DNA methylation biomarkers as potential therapeutic targets for breast cancer.

The Acer truncatum genome provides insights into nervonic acid biosynthesis.

Acer truncatum (purpleblow maple) is a woody tree species that produces seeds with high levels of valuable fatty acids (especially nervonic acid). However, the lack of a complete genome sequence has limited both basic and applied research on A. truncatum. We describe a high-quality draft genome assembly comprising 633.28 Mb (contig N50 = 773.17 kb; scaffold N50 = 46.36 Mb) with at least 28 438 predicted genes. The genome underwent an ancient triplication, similar to the core eudicots, but there have been no recent whole-genome duplication events. Acer yangbiense and A. truncatum are estimated to have diverged about 9.4 million years ago. A combined genomic, transcriptomic, metabonomic, and cell ultrastructural analysis provided new insights into the biosynthesis of very long-chain monounsaturated fatty acids. In addition, three KCS genes were found that may contribute to regulating nervonic acid biosynthesis. The KCS paralogous gene family expanded to 28 members, with 10 genes clustered together and distributed in the 0.27-Mb region of pseudochromosome 4. Our chromosome-scale genomic characterization may facilitate the discovery of agronomically important genes and stimulate functional genetic research on A. truncatum. Furthermore, the data presented also offer important foundations from which to study the molecular mechanisms influencing the production of nervonic acids.

Systematic analysis of MYB gene family in Acer rubrum and functional characterization of ArMYB89 in regulating anthocyanin biosynthesis.

The v-myb avian myeloblastosis viral oncogene homolog (MYB) family of transcription factors is extensively distributed across the plant kingdom. However, the functional significance of red maple (Acer rubrum) MYB transcription factors remains unclear. Our research identified 393 MYB transcription factors in the Acer rubrum genome, and these ArMYB members were unevenly distributed across 34 chromosomes. Among them, R2R3 was the primary MYB sub-class, which was further divided into 21 sub-groups with their Arabidopsis homologs. The evolution of the ArMYB family was also investigated, with the results revealing several R2R3-MYB sub-groups with expanded membership in woody species. Here, we report on the isolation and characterization of ArMYB89 in red maple. Quantitative real-time PCR analysis revealed that ArMYB89 expression was significantly up-regulated in red leaves in contrast to green leaves. Sub-cellular localization experiments indicated that ArMYB89 was localized in the nucleus. Further experiments revealed that ArMYB89 could interact with ArSGT1 in vitro and in vivo. Overexpression of ArMYB89 in tobacco enhances the anthocyanin content of transgenic plants. In conclusion, our results contribute to the elucidation of a theoretical basis for the ArMYB gene family, and provide a foundation for further characterization of the biological roles of MYB genes in the regulation of Acer rubrum leaf color.

Effect of the key histone modifications on the expression of genes related to breast cancer.

Abnormal histone modifications (HMs) and transcription factors (TFs) can alter the expression of cancer-related genes to promote tumorigenesis. We studied the variations of 11 HMs and 2 TFs in human breast cancer cells (MCF-7) compared to human normal mammary epithelial cells (HMEC), and the effects of HMs/TFs in various regions of the genome on the expression changes of breast cancer-related genes. Based on HMs and TFs signals' differences between MCF-7 and HMEC flanking TSSs, the up- and down-regulated genes in MCF-7 were predicted by Random Forest, and important HMs and regions were found. Results indicate that H3K79me2, H3K27ac, and H3K4me1 are particularly important for the changes of gene expression in MCF-7. Especially, H3K79me2 around the 60-th bin flanking TSSs may be the key for regulating gene expression. Our studies reveal H3K79me2 may be a core HM for breast cancer.

An energy model for recognizing the prokaryotic promoters based on molecular structure.

Promoter is an important functional elements of DNA sequences, which is in charge of gene transcription initiation. Recognizing promoter have important help for understanding the relative life phenomena. Based on the concept that promoter is mainly determined by its sequence and structure, a novel statistical physics model for predicting promoter in Escherichia coli K-12 is proposed. The total energies of DNA local structure of sequence segments in the three benchmark promoter sequence datasets, the sole prediction parameter, are calculated by using principles from statistical physics and information theory. The better results are obtained. And a web-server PhysMPrePro for predicting promoter is established at http://202.207.14.87:8032/bioinformation/PhysMPrePro/index.asp, so that other scientists can easily get their desired results by our web-server.

Revealing transcription factor and histone modification co-localization and dynamics across cell lines by integrating ChIP-seq and RNA-seq data.

BACKGROUND: Interactions among transcription factors (TFs) and histone modifications (HMs) play an important role in the precise regulation of gene expression. The context specificity of those interactions and further its dynamics in normal and disease remains largely unknown. Recent development in genomics technology enables transcription profiling by RNA-seq and protein's binding profiling by ChIP-seq. Integrative analysis of the two types of data allows us to investigate TFs and HMs interactions both from the genome co-localization and downstream target gene expression. RESULTS: We propose a integrative pipeline to explore the co-localization of 55 TFs and 11 HMs and its dynamics in human GM12878 and K562 by matched ChIP-seq and RNA-seq data from ENCODE. We classify TFs and HMs into three types based on their binding enrichment around transcription start site (TSS). Then a set of statistical indexes are proposed to characterize the TF-TF and TF-HM co-localizations. We found that Rad21, SMC3, and CTCF co-localized across five cell lines. High resolution Hi-C data in GM12878 shows that they associate most of the Hi-C peak loci with a specific CTCF-motif "anchor" and supports that CTCF, SMC3, and RAD2 co-localization serves important role in 3D chromatin structure. Meanwhile, 17 TF-TF pairs are highly dynamic between GM12878 and K562. We then build SVM models to correlate high and low expression level of target genes with TF binding and HM strength. We found that H3k9ac, H3k27ac, and three TFs (ELF1, TAF1, and POL2) are predictive with the accuracy about 85~92%. CONCLUSION: We propose a pipeline to analyze the co-localization of TF and HM and their dynamics across cell lines from ChIP-seq, and investigate their regulatory potency by RNA-seq. The integrative analysis of two level data reveals new insight for the cooperation of TFs and HMs and is helpful in understanding cell line specificity of TF/HM interactions.

Recognition of the long range enhancer-promoter interactions by further adding DNA structure properties and transcription factor binding motifs in human cell lines.

The enhancer-promoter interactions (EPIs) with strong tissue-specificity play an important role in cis-regulatory mechanism of human cell lines. However, it still remains a challenging work to predict these interactions so far. Due to that these interactions are regulated by the cooperativeness of diverse functional genomic signatures, DNA spatial structure and DNA sequence elements. In this paper, by adding DNA structure properties and transcription factor binding motifs, we presented an improved computational method to predict EPIs in human cell lines. In comparison with the results of other group on the same datasets, our best accuracies by cross-validation test were about 15%-24% higher in the same cell lines, and the accuracies by independent test were about 11%-15% higher in new cell lines. Meanwhile, we found that transcription factor binding motifs and DNA structure properties have important information that would largely determine long range EPIs prediction. From the distribution comparisons, we also found their distinct differences between interacting and non-interacting sets in each cell line. Then, the correlation analysis and network models for relationships among top-ranked functional genomic signatures indicated that diverse genomic signatures would cooperatively establish a complex regulatory network to facilitate long range EPIs. The experimental results provided additional insights about the roles of DNA intrinsic properties and functional genomic signatures in EPIs prediction.

Recognition of long-range enhancer-promoter interactions by adding genomic signatures of segmented regulatory regions.

Enhancer-promoter interaction (EPI) is an important cis-regulatory mechanism in the regulation of tissue-specific gene expression. However, it still has limitation to precisely identity these interactions so far. In this paper, using diverse genomic features for various regulatory regions, we presented a computational approach to predict EPIs with improved accuracies. Meanwhile, we comprehensively studied more potential regulatory factors that are important to EPIs prediction, such as nucleosome occupancy, enhancer RNA; and found the cell line-specificity and region-specificity of the contributions of diverse regulatory signatures. By adding genomic signatures of segmented regulatory regions, our best accuracies of cross-validation test were about 11%-16% higher than the previous results, indicating the location-specificity of genomic signatures in a regulatory region for predicting EPIs. Additionally, more training samples and related features can provide reliable performances in new cell lines. Consequently, our study provided additional insights into the roles of diverse signature features for predicting long-range EPIs.

Large-scale transcriptome comparison of sunflower genes responsive to Verticillium dahliae.

BACKGROUND: Sunflower Verticillium wilt (SVW) is a vascular disease caused by root infection with Verticillium dahliae (V. dahlia). It is a serious threat to the yield and quality of sunflower. However, chemical and agronomic measures for controlling this disease are not effective. The selection of more resistant genotypes is a desirable strategy to reduce contamination. A deeper knowledge of the molecular mechanisms and genetic basis underlying sunflower Verticillium wilt is necessary to accelerate breeding progress. RESULTS: An RNA-Seq approach was used to perform global transcriptome profiling on the roots of resistant (S18) and susceptible (P77) sunflower genotypes infected with V. dahlia. Different pairwise transcriptome comparisons were examined over a time course (6, 12 and 24 h, and 2, 3, 5 and 10 d post inoculation). In RD, SD and D datasets, 1231 genes were associated with SVW resistance in a genotype-common transcriptional pattern. Moreover, 759 and 511 genes were directly related to SVW resistance in the resistant and susceptible genotypes, respectively, in a genotype-specific transcriptional pattern. Most of the genes were demonstrated to participate in plant defense responses; these genes included peroxidase (POD), glutathione peroxidase, aquaporin PIP, chitinase, L-ascorbate oxidase, and LRR receptors. For the up-regulated genotype-specific differentially expressed genes (DEGs) in the resistant genotype, higher average fold-changes were observed in the resistant genotype compared to those in the susceptible genotype. An inverse effect was observed in the down-regulated genotype-specific DEGs in the resistant genotype. KEGG analyses showed that 98, 112 and 52 genes were classified into plant hormone signal transduction, plant-pathogen interaction and flavonoid biosynthesis categories, respectively. Many of these genes, such as CNGC, RBOH, FLS2, JAZ, MYC2 NPR1 and TGA, regulate crucial points in defense-related pathway and may contribute to V. dahliae resistance in sunflower. CONCLUSIONS: The transcriptome profiling results provided a clearer understanding of the transcripts associated with the crosstalk between sunflower and V. dahliae. The results identified several differentially expressed unigenes involved in the hyper sensitive response (HR) and the salicylic acid (SA)/jasmonic acid (JA)-mediated signal transduction pathway for resistance against V. dahliae. These results are useful for screening resistant sunflower genotypes.

Non-coding RNA identification based on topology secondary structure and reading frame in organelle genome level.

Non-coding RNA (ncRNA) genes make transcripts as same as the encoding genes, and ncRNAs directly function as RNAs rather than serve as blueprints for proteins. As the function of ncRNA is closely related to organelle genomes, it is desirable to explore ncRNA function by confirming its provenance. In this paper, the topology secondary structure, motif and the triplets under three reading frames are considered as parameters of ncRNAs. A method of SVM combining the increment of diversity (ID) algorithm is applied to construct the classifier. When the method is applied to the ncRNA dataset less than 80% sequence identity, the overall accuracies reach 95.57%, 96.40% in the five-fold cross-validation and the jackknife test, respectively. Further, for the independent testing dataset, the average prediction success rate of our method achieved 93.24%. The higher predictive success rates indicate that our method is very helpful for distinguishing ncRNAs from various organelle genomes.

De novo transcriptome sequencing of Acer palmatum and comprehensive analysis of differentially expressed genes under salt stress in two contrasting genotypes.

Maple (Acer palmatum) is an important species for landscape planting worldwide. Salt stress affects the normal growth of the Maple leaf directly, leading to loss of esthetic value. However, the limited availability of Maple genomic information has hindered research on the mechanisms underlying this tolerance. In this study, we performed comprehensive analyses of the salt tolerance in two genotypes of Maple using RNA-seq. Approximately 146.4 million paired-end reads, representing 181,769 unigenes, were obtained. The N50 length of the unigenes was 738 bp, and their total length over 102.66 Mb. 14,090 simple sequence repeats and over 500,000 single nucleotide polymorphisms were identified, which represent useful resources for marker development. Importantly, 181,769 genes were detected in at least one library, and 303 differentially expressed genes (DEGs) were identified between salt-sensitive and salt-tolerant genotypes. Among these DEGs, 125 were upregulated and 178 were downregulated genes. Two MYB-related proteins and one LEA protein were detected among the first 10 most downregulated genes. Moreover, a methyltransferase-related gene was detected among the first 10 most upregulated genes. The three most significantly enriched pathways were plant hormone signal transduction, arginine and proline metabolism, and photosynthesis. The transcriptome analysis provided a rich genetic resource for gene discovery related to salt tolerance in Maple, and in closely related species. The data will serve as an important public information platform to further our understanding of the molecular mechanisms involved in salt tolerance in Maple.

Theoretical study on the bactericidal nature of nanopatterned surfaces.

A natural biomaterial has been discovered with bactericidal activities, which is mainly attributed to its nanopatterned surface structure. The surface of Clanger cicada (Psaltoda claripennis) wings has been identified as a natural bactericidal material, which has lead to the emergence of research on the development of novel antibacterial surfaces. From the interactions between bacterial biofilms and nanopatterned surface structures, a new mechanical model is proposed that investigates the rupture of bacterial cells within the framework of the "stretching" theory. The effect of surface nanoroughness on the survival of bacterial cells is evaluated by determining the stretching ability of their cell walls. The results, calculated using Gram-positive and Gram-negative bacteria as examples, show a correlation between the stretching of the cell wall and the geometric parameters of the surface structures. The theoretical results indicate that for a given cell rigidity, the bactericidal nature of the surface is determined by the geometric parameters of the surface structures.

Sequence-specific flexibility organization of splicing flanking sequence and prediction of splice sites in the human genome.

More and more reported results of nucleosome positioning and histone modifications showed that DNA structure play a well-established role in splicing. In this study, a set of DNA geometric flexibility parameters originated from molecular dynamics (MD) simulations were introduced to discuss the structure organization around splice sites at the DNA level. The obtained profiles of specific flexibility/stiffness around splice sites indicated that the DNA physical-geometry deformation could be used as an alternative way to describe the splicing junction region. In combination with structural flexibility as discriminatory parameter, we developed a hybrid computational model for predicting potential splicing sites. And the better prediction performance was achieved when the benchmark dataset evaluated. Our results showed that the mechanical deformability character of a splice junction is closely correlated with both the splice site strength and structural information in its flanking sequences.

Irregular transcriptome reprogramming probably causes thec developmental failure of embryos produced by interspecies somatic cell nuclear transfer between the Przewalski's gazelle and the bovine.

BACKGROUND: Interspecies somatic cell nuclear transfer (iSCNT) has been regarded as a potential alternative for rescuing highly endangered species and can be used as a model for studying nuclear-cytoplasmic interactions. However, iSCNT embryos often fail to produce viable offspring. The alterations in normal molecular mechanisms contributing to extremely poor development are for the most part unknown. RESULTS: Przewalski's gazelle-bovine iSCNT embryos (PBNT) were produced by transferring Przewalski's gazelle fibroblast nuclei into enucleated bovine oocytes. The percentages of PBNT embryos that developed to morula/blastocyst stages were extremely low even with the use of various treatments that included different SCNT protocols and treatment of embryos with small molecules. Transcriptional microarray analyses of the cloned embryos showed that the upregulation of reprogramming-associated genes in bovine-bovine SCNT (BBNT) embryos was significantly higher than those observed in PBNT embryos (1527:643). In all, 139 transcripts related to various transcription regulation factors (TFs) were unsuccessfully activated in the iSCNT embryos. Maternal degradation profiles showed that 1515 genes were uniquely downregulated in the BBNT embryos, while 343 genes were downregulated in the PBNT embryos. Incompatibilities between mitochondrial DNA (mtDNA) and nuclear DNA revealed that the TOMM (translocase of outer mitochondrial membrane)/TIMM (translocase of inner mitochondrial membrane) complex-associated genes in BBNT embryos had the highest expression levels, while the PBNT embryos exhibited much lower expression rates. CONCLUSIONS: Improper degradation of maternal transcripts, incomplete activation of TFs and abnormal expression of genes associated with mitochondrial function in PBNT embryos likely contributed to incomplete reprogramming of the donor cell nuclei and therefore led to the developmental failure of these cloned embryos.

PreDNA: accurate prediction of DNA-binding sites in proteins by integrating sequence and geometric structure information.

MOTIVATION: Protein-DNA interactions often take part in various crucial processes, which are essential for cellular function. The identification of DNA-binding sites in proteins is important for understanding the molecular mechanisms of protein-DNA interaction. Thus, we have developed an improved method to predict DNA-binding sites by integrating structural alignment algorithm and support vector machine-based methods. RESULTS: Evaluated on a new non-redundant protein set with 224 chains, the method has 80.7% sensitivity and 82.9% specificity in the 5-fold cross-validation test. In addition, it predicts DNA-binding sites with 85.1% sensitivity and 85.3% specificity when tested on a dataset with 62 protein-DNA complexes. Compared with a recently published method, BindN+, our method predicts DNA-binding sites with a 7% better area under the receiver operating characteristic curve value when tested on the same dataset. Many important problems in cell biology require the dense non-linear interactions between functional modules be considered. Thus, our prediction method will be useful in detecting such complex interactions.

The effect of regions flanking target site on siRNA potency.

For a successful RNA interference (RNAi) experiment, selecting the small interference RNA (siRNA) candidates which maximize the knock down effect of the given gene is the critical step. Although various computational approaches have been attempted, the design of efficient siRNA candidates is far from satisfactory yet. In this study, we proposed a novel feature selection algorithm of combined random forest and support vector machine to predict active siRNAs. Using a publically available dataset, we demonstrated that the predictive accuracy would be markedly improved when the context sequence features outside the target site were included. The Pearson correlation coefficient for regression is as high as 0.721, compared to 0.671, 0.668, 0.680, and 0.645, for Biopredsi, i-score, ThermoComposition21 and DSIR, respectively. It revealed that siRNA-target interaction requires appropriate sequence context not only in the target site but also in a broad region flanking the target site.