Abstract Background Presently, the field of differential expression analysis of RNA-seq data is still in its infancy with new approaches constantly being proposed. Combining deep neural networks to explore gene expression information in RNA-seq data provides a novel possibility in this field. Results This study focuses on the development of a deep neural network-based RNA-seq gene expression analysis model, named Deep-cloud. Its main advantage is not only the excellent construction of deep learning network models using convolutional neural network and long short-term memory to predict the gene expression of RNA-seq, but also the more in-depth analysis of the differential gene expression between disease and normal groups of transcriptome data by combining the statistical methods of cloud model. Conclusion Comparing the results of this data analysis with those obtained by traditional differential gene analysis software such as DESeq2 and edgeR, the Deep-cloud model further improves the sensitivity of obtaining differential expressed genes. Overall, the proposed Deep-cloud paves a new pathway for the biomedical field.
RNA degradation can significantly affect the results of gene expression profiling, with subsequent analysis failing to faithfully represent the initial gene expression level. It is urgent to have an artificial intelligence approach to better utilize the limited data to obtain meaningful and reliable analysis results in the case of data with missing destination time. In this study, we propose a method based on the signal decomposition technique and deep learning, named Multi-LSTM. It is divided into two main modules: One decomposes the collected gene expression data by an empirical mode decomposition (EMD) algorithm to obtain a series of sub-modules with different frequencies to improve data stability and reduce modeling complexity. The other is based on long short-term memory (LSTM) as the core predictor, aiming to deeply explore the temporal nonlinear relationships embedded in the sub-modules. Finally, the prediction results of sub-modules are reconstructed to obtain the final prediction results of time-series transcriptomic gene expression. The results show that EMD can efficiently reduce the nonlinearity of the original data, which provides reliable theoretical support to reduce the complexity and improve the robustness of LSTM models. Overall, the decomposition-combination prediction framework can effectively predict gene expression levels at unknown time points.
Abstract Background Parkinson’s disease (PD) is the second most common neurodegenerative disease and many studies have researched its complex pathophysiological processes. However, it is unclear how PD affects the structure of transcripts in different brain regions and how changes in the transcriptomes in different brain regions affect the pathogenesis of PD. Results We generated a PD mouse model by injecting with MPTP solution. RNA sequencing was performed in the cerebral cortex, hippocampus, striatum, and cerebellum regions of the PD mouse. Compared with the control group, these four brain regions showed significant transcriptomic alterations, with the most differentially expressed genes (DEGs) found in the striatum region. The main DEGs were Lrrk2 , Mtor , Gxylt1 , C920006o11Rik , Vdac1 , Ano3 , Drd4 , and Ncan . DEGs were enriched using gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes enrichment analysis methods, which identified significant GO and molecular pathways. In addition, we used network biology methods to analyze protein–protein relationships, which can accelerate the identification of new PD drugs. The results showed that LRRK2, DRD2, IGF-1, GNAI1, GNAI3, PRKACA, PPP2R5C, and PIK3R1 played a major role in protein regulation. Conclusions Our analysis showed that these DEGs and proteins play an important role in the occurrence and development of PD. Our study also highlighted the potential use of this transcriptomic data for therapeutic strategies and treatment of PD.
Despite significant advances in parallel single-cell RNA sequencing revealing astonishing cellular heterogeneity in many tissue types, the spatial information in the tissue context remains missing.Spatial transcriptome sequencing technology is designed to distinguish the gene expression of individual cells in their original location.The technology is important for the identification of tissue function, tracking developmental processes, and pathological and molecular detection.Encoding the position information is the key to spatial transcriptomics because different methods have different encoding efficiencies and application scenarios.In this review, we focus on the latest technologies of single-cell spatial transcriptomics, including technologies based on microwell plates, barcoded bead arrays, microdissection, in situ hybridization, and barcode in situ targeting, as well as mixed separation-based technologies.Moreover, we compare these encoding methods for use as a reference when choosing the appropriate technology.
We aimed to investigate the effect of xylooligosaccharides (XOS) on the growth performances and lipid metabolism of common carp fed high-fat diets. 192 fish were randomly distributed into 24 tanks into six groups (four replicates) and were fed with control diet, high-fat diet (HFD) and HFD supplemented with 5, 10, 20 and 30 g/kgXOS respectively for 8 weeks. Fish fed HFD supplemented with 10 g/kg XOS obtained higher final body weight, weight gain, specific growth rate and protein efficiency ratio compared to those fed control diet and HFD, while feed conversion ratio showed the opposite trend. Fish fed HFD obtained higher hepatosomatic index, abdominal fat, energy intake compared to other groups, whereas the opposite was true for nitrogen retention. High plasma levels of cholesterol, triglycerides, low-density lipoprotein and low high-density lipoprotein were observed in fish fed HFD; opposite was true for fish fed HFD supplemented with 10–20 g/kg XOS. The transcription of lipoprotein lipase was up-regulated, whereas that of carnitine palmitoyltransferase I, peroxisome proliferator-activated receptors alpha, acyl-CoA oxidase and CD36 were down-regulated in fish fed HFD. Opposite trend was observed in fish fed HFD supplemented with 10–20 g/kg XOS as well as the control group. In conclusion, XOS inclusion can benefit the growth performance and lipid metabolism of common carp fed HFD.
The circRNAs sequencing results vary due to the different enrichment methods and their performance is needed to systematic comparison. This study investigated the effects of different circRNA enrichment methods on sequencing results, including abundance and species of circRNAs, as well as the sensitivity and precision. This experiment was carried out by following four common circRNA enrichment methods: including ribosomal RNA depletion (rRNA-), polyadenylation and poly (A+) RNA depletion followed by RNase R treatment (polyA+RNase R), rRNA-+polyA+RNase R and polyA+RNase R+ rRNA-. The results showed that polyA+RNase R+ rRNA - enrichment method obtained more circRNA number, higher sensitivity and abundance among them; polyA+RNase R method obtained higher precision. The linear RNAs can be thoroughly removed in all enrichment methods except rRNA depletion method. Overall, our results helps researchers to quickly selection a circRNA enrichment of suitable for own study among many enrichment methods, and it provides a benchmark framework for future improvements circRNA enrichment methods.[Figure: see text].
Background: Previous studies have shown that a large number of valuable and functional cell-free RNAs (cfRNAs) were found in follicular fluid. However, the species and characteristics of follicular fluid cfRNAs have not been reported. Furthermore, their implications are still barely understood in the evaluation of follicular fluid from follicles of different sizes, which warrants further studies. Objective: This study investigated the landscape and characteristics of follicular fluid cfRNAs, the source of organization, and the potential for distinguishing between follicles of different sizes. Methods: Twenty-four follicular fluid samples were collected from 20 patients who received in vitro fertilization (n = 9) or ICSI (n = 11), including 16 large follicular fluid and 8 small follicular fluid samples. Also, the cfRNA profile of follicular fluid samples was analyzed by RNA sequencing. Results: This result indicated that the concentration of follicular fluid cfRNAs ranged from 0.78 to 8.76 ng/ml, and fragment length was 20-200 nucleotides. The concentration and fragment length of large follicular fluid and small follicular fluid samples were not significantly different (p > 0.05). The technical replica correlation of follicular fluid samples ranged from 0.3 to 0.9, and the correlation of small follicular fluid samples was remarkably (p < 0.001) lower than that of large follicular fluid samples. Moreover, this study found that cfRNAs of the follicular fluid could be divided into 37 Ensembl RNA biotypes, and a large number of mRNAs, circRNAs, and lncRNAs were observed in the follicular fluid. The number of cfRNAs in large follicular fluid was remarkably (p < 0.05) higher than that of small follicular fluid. Furthermore, the follicular fluid contained a large amount of intact mRNA and splice junctions and a large number of tissue-derived RNAs, which are at a balanced state of supply and elimination in the follicular fluid. KEGG pathway analysis showed that differentially expressed cfRNAs were enriched in several pathways, including thyroid hormone synthesis, the cGMP-PKG signaling pathway, and inflammatory mediator regulation of TRP channels. In addition, we further showed that four cfRNAs (TK2, AHDC1, PHF21A, and TTYH1) serve as a potential indicator to distinguish the follicles of different sizes. The ROC curve shows great potential to predict follicular fluid from follicles of different sizes [area under the curve (AUC) > 0.88]. Conclusion: Overall, our study revealed that a large number of cfRNAs could be detected in follicular fluid and could serve as a potential non-invasive biomarker in distinguishing between follicles of different sizes. These results may inform the study of the utility and implementation of cfRNAs in clinical practice.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)