Machine learning reduced gene/non-coding RNA features that classify Schizophrenia patients accurately and highlight insightful gene clusters

AbstractSchizophrenia (SCZ) is a chronic and severely disabling neurodevelopmental disorder that affects people worldwide. RNA-seq has been a powerful method to detect the differentially expressed genes/non-coding RNAs in patients; however, due to overfitting problems differentially expressed targets (DETs) cannot be used properly as biomarkers. In this study, dorsolateral prefrontal cortex (dlpfc) RNA-seq data from 254 individuals’ was obtained from the CommonMind consortium and analyzed with machine learning methods, including random forest, forward feature selection (ffs), and factor analysis, to reduce the numbers of gene/non-coding RNA feature vectors to overcome overfitting problem and explore involved functional clusters. In 2-fold shuffle testing, the average predictive accuracy for SCZ patients was 67% based on coding genes, and the 96% based on long non-coding RNAs (lncRNAs). Coding genes were further clustered into 14 factors and lncRNAs were clustered into 45 factors to represent the underlying features. The largest contribution factor for coding genes contains number of genes critical in neurodevelopment and previously reported in relation with various brain disorders. Genomic loci of lncRNAs were more insightful, enriched for genes critical in synapse function (p=7.3E-3), cell junction (p=0.017), neuron differentiation (p=8.3E-3), phosphorylation (8.2E-4), and involving the Wnt signaling pathway (p=0.029). Taken together, machine learning is a powerful algorithm to reduce functional biomarkers in SCZ patients. The lncRNAs capture the characteristics of SCZ tissue more accurately than mRNA as the formers regulate every level of gene expression, not limited to mRNA levels.

Download Full-text

Machine Learning Reduced Gene/Non-Coding RNA Features That Classify Schizophrenia Patients Accurately and Highlight Insightful Gene Clusters

International Journal of Molecular Sciences ◽

10.3390/ijms22073364 ◽

2021 ◽

Vol 22 (7) ◽

pp. 3364

Author(s):

Yichuan Liu ◽

Hui-Qi Qu ◽

Xiao Chang ◽

Lifeng Tian ◽

Jingchun Qu ◽

...

Keyword(s):

Machine Learning ◽

Dorsolateral Prefrontal Cortex ◽

Predictive Accuracy ◽

Gene Clusters ◽

Differentially Expressed ◽

Mrna Levels ◽

Rna Seq ◽

Non Coding Rna ◽

Dorsolateral Prefrontal ◽

Non Coding Rnas

RNA-seq has been a powerful method to detect the differentially expressed genes/long non-coding RNAs (lncRNAs) in schizophrenia (SCZ) patients; however, due to overfitting problems differentially expressed targets (DETs) cannot be used properly as biomarkers. This study used machine learning to reduce gene/non-coding RNA features. Dorsolateral prefrontal cortex (dlpfc) RNA-seq data from 254 individuals was obtained from the CommonMind consortium. The average predictive accuracy for SCZ patients was 67% based on coding genes, and 96% based on long non-coding RNAs (lncRNAs). Machine learning is a powerful algorithm to reduce functional biomarkers in SCZ patients. The lncRNAs capture the characteristics of SCZ tissue more accurately than mRNA as the former regulate every level of gene expression, not limited to mRNA levels.

Download Full-text

RNA-Seq Reveals the Expression Profiles of Long Non-Coding RNAs in Lactating Mammary Gland from Two Sheep Breeds with Divergent Milk Phenotype

Animals ◽

10.3390/ani10091565 ◽

2020 ◽

Vol 10 (9) ◽

pp. 1565

Author(s):

Zhiyun Hao ◽

Yuzhu Luo ◽

Jiqing Wang ◽

Jiang Hu ◽

Xiu Liu ◽

...

Keyword(s):

Mammary Gland ◽

Signaling Pathway ◽

Target Genes ◽

Mammary Epithelial Cells ◽

Expression Profiles ◽

Wnt Signaling Pathway ◽

Mammary Epithelial ◽

Differentially Expressed ◽

Rna Seq ◽

Non Coding Rnas

Long non-coding RNAs (lncRNAs) are a kind of non-coding RNA with >200 nucleotides in length. Some lncRNAs have been proven to have clear regulatory functions in many biological processes of mammals. However, there have been no reports on the roles of lncRNAs in ovine mammary gland tissues. In the study, the expression profiles of lncRNAs were studied using RNA-Seq in mammary gland tissues from lactating Small-Tailed Han (STH) ewes and Gansu Alpine Merino (GAM) ewes with different milk yield and ingredients. A total of 1894 lncRNAs were found to be expressed. Compared with the GAM ewes, the expression levels of 31 lncRNAs were significantly up-regulated in the mammary gland tissues of STH ewes, while 37 lncRNAs were remarkably down-regulated. Gene Ontogeny (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis found that the target genes of differentially expressed lncRNAs were enriched in the development and proliferation of mammary epithelial cells, morphogenesis of mammary gland, ErbB signaling pathway, and Wnt signaling pathway. Some miRNA sponges of differentially expressed lncRNAs, reported to be associated with lactation and mammary gland morphogenesis, were found in a lncRNA-miRNA network. This study reveals comprehensive lncRNAs expression profiles in ovine mammary gland tissues, thereby providing a further understanding of the functions of lncRNAs in the lactation and mammary gland development of sheep.

Download Full-text

Effects of Castration on miRNA, lncRNA, and mRNA Profiles in Mice Thymus

Genes ◽

10.3390/genes11020147 ◽

2020 ◽

Vol 11 (2) ◽

pp. 147

Author(s):

Bingxin Li ◽

Kaizhao Zhang ◽

Yaqiong Ye ◽

Jingjing Xing ◽

Yingying Wu ◽

...

Keyword(s):

Gene Networks ◽

Gonadal Hormones ◽

Differentially Expressed ◽

Thymic Epithelial Cells ◽

Mrna Levels ◽

Rna Seq ◽

Thymic Development ◽

Promoter Sequences ◽

Differentially Expressed Mirnas ◽

Non Coding Rnas

Thymic degeneration and regeneration are regulated by estrogen and androgen. Recent studies have found that long non-coding RNAs (lncRNAs) and microRNAs (miRNAs) are involved in organ development. In this study, RNA sequencing (RNA-seq) results showed that ovariectomy significantly affected 333 lncRNAs, 51 miRNAs, and 144 mRNAs levels (p < 0.05 and |log2fold change| > 1), and orchiectomy significantly affected 165 lncRNAs, 165 miRNAs, and 208 mRNA levels in the thymus. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis showed that differentially expressed genes (DEGs) were closely related to cell development and immunity. Next, we constructed two lncRNA–miRNA–mRNA networks using Cytoscape based on the targeting relationship between differentially expressed miRNAs (DEMs) and DEGs and differentially expressed lncRNAs (DELs) analyzed by TargetScan and miRanda. Besides, we screened DEGs that were significantly enriched in GO and in ceRNA networks to verify their expression in thymocytes and thymic epithelial cells (TECs). In addition, we analyzed the promoter sequences of DEGs, and identified 25 causal transcription factors. Finally, we constructed transcription factor-miRNA-joint target gene networks. In conclusion, this study reveals the effects of estrogen and androgen on the expression of miRNAs, lncRNAs, and mRNAs in mice thymus, providing new insights into the regulation of thymic development by gonadal hormones and non-coding RNAs.

Download Full-text

Long Intergenic Non-Coding RNAs in the Mammary Parenchyma and Fat Pad of Pre-Weaning Heifer Calves: Identification and Functional Analysis

Animals ◽

10.3390/ani11051268 ◽

2021 ◽

Vol 11 (5) ◽

pp. 1268

Author(s):

Shengchao Zhang ◽

Sibtain Ahmad ◽

Yuxia Zhang ◽

Guohua Hua ◽

Jianming Yi

Keyword(s):

Mammary Gland ◽

Crude Protein ◽

Regulatory Mechanism ◽

Differentially Expressed ◽

Milk Replacer ◽

Rna Seq ◽

Fat Pad ◽

Mammary Fat Pad ◽

Non Coding Rnas ◽

Gland Development

Enhanced plane of nutrition at pre-weaning stage can promote the development of mammary gland especially heifer calves. Although several genes are involved in this process, long intergenic non-coding RNAs (lincRNAs) are regarded as key regulators in the regulated network and are still largely unknown. We identified and characterized 534 putative lincRNAs based on the published RNA-seq data, including heifer calves in two groups: fed enhanced milk replacer (EH, 1.13 kg/day, including 28% crude protein, 25% fat) group and fed restricted milk replacer (R, 0.45 kg/day, including 20% crude protein, 20% fat) group. Sub-samples from the mammary parenchyma (PAR) and mammary fat pad (MFP) were harvested from heifer calves. According to the information of these lincRNAs’ quantitative trait loci (QTLs), the neighboring and co-expression genes were used to predict their function. By comparing EH vs R, 79 lincRNAs (61 upregulated, 18 downregulated) and 86 lincRNAs (54 upregulated, 32 downregulated) were differentially expressed in MFP and PAR, respectively. In MFP, some differentially expressed lincRNAs (DELs) are involved in lipid metabolism pathways, while, in PAR, among of DELs are involved in cell proliferation pathways. Taken together, this study explored the potential regulatory mechanism of lincRNAs in the mammary gland development of calves under different planes of nutrition.

Download Full-text

Transcriptomic and ChIP-seq Integrative Analysis Reveals Important Roles of Epigenetically Regulated lncRNAs in Placental Development in Meishan Pigs

Genes ◽

10.3390/genes11040397 ◽

2020 ◽

Vol 11 (4) ◽

pp. 397

Author(s):

Dadong Deng ◽

Xihong Tan ◽

Kun Han ◽

Ruimin Ren ◽

Jianhua Cao ◽

...

Keyword(s):

Differentially Expressed ◽

Rna Seq ◽

Sequencing Data ◽

Placental Development ◽

Cytoskeleton Organization ◽

New Class ◽

Chromatin Immunoprecipitation Sequencing ◽

Non Coding Rnas ◽

Two Stages ◽

Regulatory Functions

The development of the placental fold, which increases the maternal–fetal interacting surface area, is of primary importance for the growth of the fetus throughout the whole pregnancy. However, the mechanisms involved remain to be fully elucidated. Increasing evidence has revealed that long non-coding RNAs (lncRNAs) are a new class of RNAs with regulatory functions and could be epigenetically regulated by histone modifications. In this study, 141 lncRNAs (including 73 up-regulated and 68 down-regulated lncRNAs) were identified to be differentially expressed in the placentas of pigs during the establishment and expanding stages of placental fold development. The differentially expressed lncRNAs and genes (DElncRNA-DEgene) co-expression network analysis revealed that these differentially expressed lncRNAs (DElncRNAs) were mainly enriched in pathways of cell adhesion, cytoskeleton organization, epithelial cell differentiation and angiogenesis, indicating that the DElncRNAs are related to the major events that occur during placental fold development. In addition, we integrated the RNA-seq (RNA sequencing) data with the ChIP-seq (chromatin immunoprecipitation sequencing) data of H3K4me3/H3K27ac produced from the placental samples of pigs from the two stages (gestational days 50 and 95). The analysis revealed that the changes in H3K4me3 and/or H3K27ac levels were significantly associated with the changes in the expression levels of 37 DElncRNAs. Furthermore, several H3K4me3/H3K27ac-lncRNAs were characterized to be significantly correlated with genes functionally related to placental development. Thus, this study provides new insights into understanding the mechanisms for the placental development of pigs.

Download Full-text

Abstract 286: Long Noncoding RNAs Are Differentially Expressed in Heart Failure

Circulation Research ◽

10.1161/res.111.suppl_1.a286 ◽

2012 ◽

Vol 111 (suppl_1) ◽

Author(s):

Emma L Robinson ◽

Syed Haider ◽

Hillary Hei ◽

Richard T Lee ◽

Roger S Foo

Keyword(s):

Gene Expression ◽

Heart Failure ◽

Significant Proportion ◽

Differentially Expressed ◽

Rna Seq ◽

Control Of Gene Expression ◽

Failing Heart ◽

Novel Transcripts ◽

Non Coding Rnas

Heart failure comprises of clinically distinct inciting causes but a consistent pattern of change in myocardial gene expression supports the hypothesis that unifying biochemical mechanisms underlie disease progression. The recent RNA-seq revolution has enabled whole transcriptome profiling, using deep-sequencing technologies. Up to 70% of the genome is now known to be transcribed into RNA, a significant proportion of which is long non-coding RNAs (lncRNAs), defined as polyribonucleotides of ≥200 nucleotides. This project aims to discover whether the myocardium expression of lncRNAs changes in the failing heart. Paired end RNA-seq from a 300-400bp library of ‘stretched’ mouse myocyte total RNA was carried out to generate 76-mer sequence reads. Mechanically stretching myocytes with equibiaxial stretch apparatus mimics pathological hypertrophy in the heart. Transcripts were assembled and aligned to reference genome mm9 (UCSC), abundance determined and differential expression of novel transcripts and alternative splice variants were compared with that of control (non-stretched) mouse myocytes. Five novel transcripts have been identified in our RNA-seq that are differentially expressed in stretched myocytes compared with non-stretched. These are regions of the genome that are currently unannotated and potentially are transcribed into non-coding RNAs. Roles of known lncRNAs include control of gene expression, either by direct interaction with complementary regions of the genome or association with chromatin remodelling complexes which act on the epigenome.Changes in expression of genes which contribute to the deterioration of the failing heart could be due to the actions of these novel lncRNAs, immediately suggesting a target for new pharmaceuticals. Changes in the expression of these novel transcripts will be validated in a larger sample size of stretched myocytes vs non-stretched myocytes as well as in the hearts of transverse aortic constriction (TAC) mice vs Sham (surgical procedure without the aortic banding). In vivo investigations will then be carried out, using siLNA antisense technology to silence novel lncRNAs in mice.

Download Full-text

Long-non Coding RNAs Related to Fat Deposition in Pigs Included lncRNA Corresponding to Human MALAT1

10.20944/preprints202103.0356.v1 ◽

2021 ◽

Author(s):

Katarzyna Piórkowska ◽

Kacper Żukowski ◽

Katarzyna Ropka-Molik ◽

Mirosław Tyra

Keyword(s):

Regulatory Elements ◽

Fat Deposition ◽

Differentially Expressed ◽

Rna Molecules ◽

Non Coding Rna ◽

Sequencing Method ◽

Non Coding Rnas ◽

Potential Interactions ◽

Long Non Coding Rna ◽

Generation Sequencing

Obesity is a problem in the last decades since the development of different technologies forced the submission of a faster pace of life, resulting in nutrition style changes. In turn, domestic pigs are an excellent animal model in recognition of adiposity-related processes, corresponding to the size of individual organs, the distribution of body fat in the organism, and similar metabolism. The present study applied the next-generation sequencing method to identify adipose tissue (AT) transcriptomic signals related to increased fat content by identifying differentially expressed genes (DEGs), included long-non coding RNA molecules. The Freiburg RNA tool was applied to recognise predicting hybridisation energy of RNA-RNA interactions. The results indicated several long non-coding RNAs (lncRNAs) whose expression was significantly positively or negatively associated with fat deposition. lncRNAs play an essential role in regulating gene expression by sponging miRNA, binding transcripts, facilitating translation, or coding other smaller RNA regulatory elements. In the pig fat tissue of obese group, increased expression of lncRNAs corresponding to human MALAT1 was observed that previously recognised in the obesity-related context. Moreover, hybridisation energy analyses pinpointed numerous potential interactions between identified differentially expressed lncRNAs, and obesity-related genes and miRNAs expressed in AT.

Download Full-text

Whole Blood Transcriptome Analysis in Children with Sickle Cell Anemia

Frontiers in Genetics ◽

10.3389/fgene.2021.737741 ◽

2022 ◽

Vol 12 ◽

Author(s):

Beatrice E. Gee ◽

Andrea Pearson ◽

Iris Buchanan-Perry ◽

Roger P. Simon ◽

David R. Archer ◽

...

Keyword(s):

Gene Expression ◽

Rna Sequencing ◽

Sickle Cell ◽

Sickle Cell Anemia ◽

Whole Blood ◽

Differentially Expressed ◽

Mrna Levels ◽

Control Subjects ◽

Non Coding Rna ◽

And Control

Whole transcriptome RNA-sequencing was performed to quantify RNA expression changes in whole blood samples collected from steady state sickle cell anemia (SCA) and control subjects. Pediatric SCA and control subjects were recruited from Atlanta (GA)—based hospital(s) systems and consented for RNA sequencing. RNA sequencing was performed on an Ion Torrent S5 sequencer, using the Ion Total RNA-seq v2 protocol. Data were aligned to the hg19 reference genome and analyzed in the Partek Genomics studio package (v7.0). 223 genes were differentially expressed between SCA and controls (± 1.5 fold change FDR p < 0.001) and 441 genes show differential transcript expression (± 1.5 fold FDR p < 0.001). Differentially expressed RNA are enriched for hemoglobin associated genes and ubiquitin-proteasome pathway genes. Further analysis shows higher gamma globin gene expression in SCA (33-fold HBG1 and 49-fold HBG2, both FDR p < 0.05), which did not correlate with hemoglobin F protein levels. eQTL analysis identified SNPs in novel non-coding RNA RYR2 gene as having a potential regulatory role in HBG1 and HBG2 expression levels. Gene expression correlation identified JHDM1D-AS1(KDM7A-DT), a non-coding RNA associated with angiogenesis, enhanced GATA1 and decreased JAK-STAT signaling to correlate with HBG1 and HBG2 mRNA levels. These data suggest novel regulatory mechanisms for fetal hemoglobin regulation, which may offer innovative therapeutic approaches for SCA.

Download Full-text

Whole Transcriptome Analysis: Implication to Estrous Cycle Regulation

10.21203/rs.3.rs-292826/v1 ◽

2021 ◽

Author(s):

Xiaopeng An ◽

Yue Zhang ◽

Fu Li ◽

Zhanhang Wang ◽

Shaohua Yang ◽

...

Keyword(s):

Estrous Cycle ◽

Transcriptome Analysis ◽

Circular Rna ◽

Differentially Expressed ◽

Non Coding Rna ◽

Non Coding Rnas ◽

Whole Transcriptome Analysis ◽

Cycle Regulation ◽

Whole Transcriptome ◽

Goat Ovary

Abstract BackgroundEstrous cycle is one of female characteristics after sexual maturity, including estrus (ES) and diestrus (DS) stages. Estrous cycle is important in female physiology and its disorder may lead to diseases. In the latest years, effects of non-coding RNAs and mRNA on estrous cycle start to arouse much concern, however, a whole transcriptome analysis among non-coding RNAs and mRNA has not been reported.ResultsHere we report a whole transcriptome analysis of goat ovary in estrus and diestrus periods. Estrus synchronization was conducted to induce the estrus phase and on day 32, the goats naturally shifted into diestrus stage. The ovary RNA of estrus and diestrus stages was respectively collected to perform RNA-sequencing. Then the circular RNA; microRNA; long non-coding RNA; mRNA databases of goat ovary were acquired, and the differentially expressions between estrus and diestrus stages were screened to construct circRNA-miRNA-mRNA/lncRNA and lncRNA-miRNA/mRNA networks, thus providing potential pathways that involved in the regulation of estrous cycle. Differentially expressed mRNAs, such as MMP9, TIMP1, 3BHSD and PTGIS, and differentially expressed microRNAs, such as miR-21-3p，miR-202-3p and miR-223-3p, which play key roles in estrous cycle regulation were extracted from the network.ConclusionsOur data provided the miRNA, circRNA, lncRNA and mRNA databases of goat ovary and each differentially expressed profile between ES and DS. Networks among differentially expressed miRNAs, circRNAs, lncRNAs and mRNAs were constructed to provide valuable resources for the study of estrous cycle and related diseases.

Download Full-text

Identification of Novel lncRNAs Differentially Expressed in Placentas of Chinese Ningqiang Pony and Yili Horse Breeds

Animals ◽

10.3390/ani10010119 ◽

2020 ◽

Vol 10 (1) ◽

pp. 119

Author(s):

Yabin Pu ◽

Yanli Zhang ◽

Tian Zhang ◽

Jianlin Han ◽

Yuehui Ma ◽

...

Keyword(s):

Body Size ◽

Limb Development ◽

Small Body ◽

Expression Patterns ◽

Differentially Expressed ◽

Myoblast Differentiation ◽

Rna Seq ◽

Skeletal Myoblast ◽

Control Procedures ◽

Non Coding Rnas

As a nutrient sensor, the placenta plays a key role in regulating fetus growth and development. Long non-coding RNAs (lncRNAs) have been shown to regulate growth-related traits. However, the biological function of lncRNAs in horse placentas remains unclear. To compare the expression patterns of lncRNAs in the placentas of the Chinese Ningqiang (NQ) and Yili (YL) breeds, we performed a transcriptome analysis using RNA sequencing (RNA-seq) technology. NQ is a pony breed with an average adult height at the withers of less than 106 cm, whereas that of YL is around 148 cm. Based on 813 million high-quality reads and stringent quality control procedures, 3011 transcripts coding for 1464 placental lncRNAs were identified and mapped to the horse reference genome. We found 107 differentially expressed lncRNAs (DELs) between NQ and YL, including 68 up-regulated and 39 down-regulated DELs in YL. Six (TBX3, CACNA1F, EDN3, KAT5, ZNF281, TMED2, and TGFB1) out of the 233 genes targeted by DELs were identified as being involved in limb development, skeletal myoblast differentiation, and embryo development. Two DELs were predicted to target the TBX3 gene, which was found to be under strong selection and associated with small body size in the Chinese Debao pony breed. This finding suggests the potential functional significance of placental lncRNAs in regulating horse body size.

Download Full-text