Search

Goat milk is abundant in nutrients, particularly in milk fats, which confer health benefits to humans. Exploring the regulatory mechanism of fatty acid synthesis is highly important to understand milk composition manipulation. In this study, we used chromatin immunoprecipitation sequencing (ChIP-seq) on goat mammary glands at different lactation stages which revealed a novel lactation regulatory factor: cell death-inducing DFFA-like effector B (CIDEB). RT-qPCR results revealed that CIDEB was significantly upregulated during lactation in dairy goats. CIDEB overexpression significantly increased the expression levels of genes involved in fatty acid synthesis ( ACACA , SCD1 , p < 0.05; ELOVL6 , p < 0.01), lipid droplet formation ( XDH , p < 0.05), and triacylglycerol (TAG) synthesis ( DGAT1 , p < 0.05; GPAM , p < 0.01) in goat mammary epithelial cells (GMECs). The contents of lipid droplets, TAG, and cholesterol were increased ( p < 0.05) in CIDEB-overexpressing GMECs, and knockdown of CIDEB led to the opposite results. In addition, CIDEB knockdown significantly decreased the proportion of C16:0 and total C18:2. Luciferase reporter assays indicated that X-box binding protein 1 (XBP1) promoted CIDEB transcription via XBP1 binding sites located in the CIDEB promoter. Furthermore, CIDEB knockdown attenuated the stimulatory effect of XBP1 on lipid droplet accumulation. Collectively, these findings elucidate the critical regulatory roles of CIDEB in milk fat synthesis, thus providing new insights into improving the quality of goat milk.

Prediction of binding sites for transcription factors is important to understand how the latter regulate gene expression and how this regulation can be modulated for therapeutic purposes. A consistent number of references address this issue with different approaches, Machine Learning being one of the most successful. Nevertheless, we note that many such approaches fail to propose a robust and meaningful method to embed the genetic data under analysis. We try to overcome this problem by proposing a bidirectional transformer-based encoder, empowered by bidirectional long-short term memory layers and with a capsule layer responsible for the final prediction. To evaluate the efficiency of the proposed approach, we use benchmark ChIP-seq datasets of five cell lines available in the ENCODE repository (A549, GM12878, Hep-G2, H1-hESC, and Hela). The results show that the proposed method can predict TFBS within the five different cell lines very well; moreover, cross-cell predictions provide satisfactory results as well. Experiments conducted across cell lines are reinforced by the analysis of five additional lines used only to test the model trained using the others. The results confirm that prediction across cell lines remains very high, allowing an extensive cross-transcription factor analysis to be performed from which several indications of interest for molecular biology may be drawn.

Recent advances in single-cell sequencing assays for the transposase-accessibility chromatin (scATAC-seq) technique have provided cell-specific chromatin accessibility landscapes of cis-regulatory elements, providing deeper insights into cellular states and dynamics. However, few research efforts have been dedicated to modeling the relationship between regulatory grammars and single-cell chromatin accessibility and incorporating different analysis scenarios of scATAC-seq data into the general framework. To this end, we propose a unified deep learning framework based on the ProdDep Transformer Encoder, dubbed PROTRAIT, for scATAC-seq data analysis. Specifically motivated by the deep language model, PROTRAIT leverages the ProdDep Transformer Encoder to capture the syntax of transcription factor (TF)-DNA binding motifs from scATAC-seq peaks for predicting single-cell chromatin accessibility and learning single-cell embedding. Based on cell embedding, PROTRAIT annotates cell types using the Louvain algorithm. Furthermore, according to the identified likely noises of raw scATAC-seq data, PROTRAIT denoises these values based on predated chromatin accessibility. In addition, PROTRAIT employs differential accessibility analysis to infer TF activity at single-cell and single-nucleotide resolution. Extensive experiments based on the Buenrostro2018 dataset validate the effeteness of PROTRAIT for chromatin accessibility prediction, cell type annotation, and scATAC-seq data denoising, therein outperforming current approaches in terms of different evaluation metrics. Besides, we confirm the consistency between the inferred TF activity and the literature review. We also demonstrate the scalability of PROTRAIT to analyze datasets containing over one million cells.

Transcription factors (TFs) bind DNA in a sequence-specific manner and are generally cell type-specific factors and/or developmental master regulators. In contrast, general TFs (GTFs) are part of very large protein complexes and serve for RNA polymerases' recruitment to promoter sequences, generally in a cell type-independent manner. Whereas, several TFs have been proven to serve as anchors for the 3D genome organization, the role of GTFs in genome architecture have not been carefully explored. Here, we used ChIP-seq and Hi-C data to depict the role of TFIIIC, one of the RNA polymerase III GTFs, in 3D genome organization. We find that TFIIIC genome occupancy mainly occurs at specific regions, which largely correspond to Alu elements; other characteristic classes of repetitive elements (REs) such as MIR, FLAM-C and ALR/alpha are also found depending on the cell's developmental origin. The analysis also shows that TFIIIC-enriched regions are involved in cell type-specific DNA looping, which does not depend on colocalization with the master architectural protein CTCF. This work extends previous knowledge on the role of TFIIIC as a bona fide genome organizer whose action participates in cell type-dependent 3D genome looping via binding to REs.

The explosive development of next-generation sequencing-based technologies has allowed us to take an unprecedented look at many molecular signatures of the non-coding genome. In particular, the ChIP-seq (Chromatin ImmunoPrecipitation followed by sequencing) technique is now very commonly used to assess the proteins associated with different non-coding DNA regions genome-wide. While the analysis of such data related to transcription factor binding is relatively straightforward, many modified histone variants, such as H3K27me3, are very important for the process of gene regulation but are very difficult to interpret. We propose a novel method, called HERON (HiddEn MaRkov mOdel based peak calliNg), for genome-wide data analysis that is able to detect DNA regions enriched for a certain feature, even in difficult settings of weakly enriched long DNA domains. We demonstrate the performance of our method both on simulated and experimental data.

Currently, the detection of the allele asymmetry of gene expression from RNA-seq data or the transcription factor binding from ChIP-seq data is one of the approaches used to identify the functional genetic variants that can affect gene expression (regulatory SNPs or rSNPs). In this study, we searched for rSNPs using the data for human pulmonary arterial endothelial cells (PAECs) available from the Sequence Read Archive (SRA). Allele-asymmetric binding and expression events are analyzed in paired ChIP-seq data for H3K4me3 mark and RNA-seq data obtained for 19 individuals. Two statistical approaches, weighted z-scores and predicted probabilities, were used to improve the efficiency of finding rSNPs. In total, we identified 14,266 rSNPs associated with both allele-specific binding and expression. Among them, 645 rSNPs were associated with GWAS phenotypes; 4746 rSNPs were reported as eQTLs by GTEx, and 11,536 rSNPs were located in 374 candidate transcription factor binding motifs. Additionally, we searched for the rSNPs associated with gene expression using an SRA RNA-seq dataset for 281 clinically annotated human postmortem brain samples and detected eQTLs for 2505 rSNPs. Based on these results, we conducted Gene Ontology (GO), Disease Ontology (DO), and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses and constructed the protein-protein interaction networks to represent the top-ranked biological processes with a possible contribution to the phenotypic outcome.