Ultra-high throughput sequencing-based small RNA discovery and discrete statistical biomarker analysis in a collection of cervical tumours and matched controls

Interplay of miR-542, miR-126, miR-143 and miR-26b with PI3K-Akt is a Diagnostic Signal and Putative Regulatory Target in HPV-Positive Cervical Cancer

Human papillomavirus accounts for 99.7% of all cervical cancer cases worldwide. The viral oncoproteins alter normal cell signaling and gene expression, resulting in loss of cell cycle control and cancer development. Also, microRNAs (miRNAs) have been reported to play a critical role in cervical carcinogenesis. Especially these are not only appropriate targets for therapeutic intervention in cervical cancer but also early diagnostic signals. The given study tries to improve the sparse knowledge on miRNAs and their role in this physiological context. Deregulated miRNAs were identified by analyzing the raw data of the well-founded GSE20592 dataset including 16 tumor/normal pairs of human cervical tissue samples. The dataset was quantified by a conservative strategy based on HTSeq and Salmon, followed by target prediction via TargetScan and miRDB. The comprehensive pathway analysis of all factors was performed using DAVID. The theoretical results were subject of a stringent experimental validation in a well-characterized clinical cohort of 30 tumor/normal pairs of cervical samples. The top 31 miRNAs and their 140 primary target genes were closely intertwined with the PI3K-Akt signaling pathway. MiR-21-3p and miR-1-3p showed a prominent regulatory role while miR-542, miR-126, miR-143, and miR-26b are directly targeting both PI3K and AKT. This study provides insights into the regulation of PI3K-Akt signaling as an important inducer of cervical cancer and identified miR-542, miR-126, miR-143, and miR-26b as promising inhibitors of the PI3K-Akt action.

Machine Learning Made Easy (MLme): a comprehensive toolkit for machine learning-driven data analysis

BACKGROUND

Machine learning (ML) has emerged as a vital asset for researchers to analyze and extract valuable information from complex datasets. However, developing an effective and robust ML pipeline can present a real challenge, demanding considerable time and effort, thereby impeding research progress. Existing tools in this landscape require a profound understanding of ML principles and programming skills. Furthermore, users are required to engage in the comprehensive configuration of their ML pipeline to obtain optimal performance.

RESULTS

To address these challenges, we have developed a novel tool called Machine Learning Made Easy (MLme) that streamlines the use of ML in research, specifically focusing on classification problems at present. By integrating 4 essential functionalities-namely, Data Exploration, AutoML, CustomML, and Visualization-MLme fulfills the diverse requirements of researchers while eliminating the need for extensive coding efforts. To demonstrate the applicability of MLme, we conducted rigorous testing on 6 distinct datasets, each presenting unique characteristics and challenges. Our results consistently showed promising performance across different datasets, reaffirming the versatility and effectiveness of the tool. Additionally, by utilizing MLme's feature selection functionality, we successfully identified significant markers for CD8+ naive (BACH2), CD16+ (CD16), and CD14+ (VCAN) cell populations.

CONCLUSION

MLme serves as a valuable resource for leveraging ML to facilitate insightful data analysis and enhance research outcomes, while alleviating concerns related to complex coding scripts. The source code and a detailed tutorial for MLme are available at https://github.com/FunctionalUrology/MLme.

NAxtra magnetic nanoparticles for low-cost, efficient isolation of mammalian DNA and RNA

A cost-effective, viral nucleic acid (NA) isolation kit based on NAxtra magnetic nanoparticles was developed at the Norwegian University of Science and Technology in response to the shortage of commercial kits for isolation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA during the coronavirus disease 2019 (COVID-19) pandemic. This method showed comparable sensitivity to available kits at significantly reduced cost, making its application for other biological sources an intriguing prospect. Thus, based on this low-cost nucleic acid extraction technology, we developed a simple, low- and high-throughput, efficient method for isolation of high-integrity total NA, DNA and RNA from mammalian cell lines (monolayer) and organoids (3D-cultures). The extracted NA are compatible with downstream applications including (RT-)qPCR and next-generation sequencing. When automated, NA isolation can be performed in 14 min for up to 96 samples, yielding similar quantities to available kits.

Machine Learning Made Easy (MLme): A Comprehensive Toolkit for Machine Learning-Driven Data Analysis

Background

Machine learning (ML) has emerged as a vital asset for researchers to analyze and extract valuable information from complex datasets. However, developing an effective and robust ML pipeline can present a real challenge, demanding considerable time and effort, thereby impeding research progress. Existing tools in this landscape require a profound understanding of ML principles and programming skills. Furthermore, users are required to engage in the comprehensive configuration of their ML pipeline to obtain optimal performance.

Results

To address these challenges, we have developed a novel tool called Machine Learning Made Easy (MLme) that streamlines the use of ML in research, specifically focusing on classification problems at present. By integrating four essential functionalities, namely Data Exploration, AutoML, CustomML, and Visualization, MLme fulfills the diverse requirements of researchers while eliminating the need for extensive coding efforts. To demonstrate the applicability of MLme, we conducted rigorous testing on six distinct datasets, each presenting unique characteristics and challenges. Our results consistently showed promising performance across different datasets, reaffirming the versatility and effectiveness of the tool. Additionally, by utilizing MLme's feature selection functionality, we successfully identified significant markers for CD8+ naive (BACH2), CD16+ (CD16), and CD14+ (VCAN) cell populations.

Conclusion

MLme serves as a valuable resource for leveraging machine learning (ML) to facilitate insightful data analysis and enhance research outcomes, while alleviating concerns related to complex coding scripts. The source code and a detailed tutorial for MLme are available at https://github.com/FunctionalUrology/MLme.

miRNAs signature as potential biomarkers for cervical precancerous lesions in human papillomavirus positive women

Biomarkers to identify women at risk of cervical cancer among those with high-risk HPV infection (hrHPV+) are needed. Deregulated expression of microRNAs (miRNAs) contributes to hrHPV-induced cervical carcinogenesis. We aimed at identifying miRNAs with the capacity to distinguish high (CIN2+) and low (≤ CIN1) grade cervical lesions. We sequenced miRNA libraries from Formalin-Fixed Paraffin-Embedded (FFPE) tissues from women with CIN2+ (n = 10) and age-matched women with ≤ CIN1 (n = 10), randomly and retrospectively selected from a trial that followed women for 24 months after a hrHPV+ test at the screening visit. Five miRNAs differentially expressed were validated by RT-qPCR in an independent set of FFPE tissues with a reviewed diagnosis of CIN2+ (n = 105) and ≤ CIN1 (n = 105). The Ingenuity Pathway Analysis (IPA) was conducted to identify mRNAs inversely correlated with the top 25 differentially expressed miRNAs. Inverse correlations with 401 unique mRNA targets were identified for fourteen of the top 25 differentially expressed miRNAs. Eleven of these miRNAs targeted 26 proteins of pathways deregulated by HPV E6 and E7 oncoproteins and two of them, miR-143-5p and miR-29a-3p, predicted CIN2+ and CIN3+ in the independent validation by RT-qPCR of FFPE tissues from hrHPV-positive women.

The ceramide synthase (CERS/LASS) family: Functions involved in cancer progression

INTRODUCTION

Ceramide synthases (CERSes) are also known longevity assurance (LASS) genes. CERSes play important roles in the regulation of cancer progression. The CERS family is expressed in a variety of human tumours and is involved in tumorigenesis. They are closely associated with the progression of liver, breast, cervical, ovarian, colorectal, head and neck squamous cell, gastric, lung, prostate, oesophageal, pancreatic and blood cancers. CERSes play diverse and important roles in the regulation of cell survival, proliferation, apoptosis, migration, invasion, and drug resistance. The differential expression of CERSes in tumour and nontumour cells and survival analysis of cancer patients suggest that some CERSes could be used as potential prognostic markers. They are also important potential targets for cancer therapy.

METHODS

In this review, we summarize the available evidence on the inhibitory or promotive roles of CERSes in the progression of many cancers. Furthermore, we summarize the identified upstream and downstream molecular mechanisms that may regulate the function of CERSes in cancer settings.

Screening of Precancerous Lesions in Women with Human Papillomavirus (HPV) Infection by Molecular Typing and MicroRNA Analysis

Human papillomavirus (HPV) is causatively associated with cervical cancer, the fourth most common malignant disease of women worldwide: (1) The aim of the proposed study is to implement routine diagnostics of HPV precancerous cervical lesions by introducing new molecular diagnostic tools. (2) Methods: This is a retrospective cohort study with a total of twenty-two formalin-fixed paraffin-embedded (FFPE) cervical samples of various sample type (nine biopsy and thirteen conization) each patient had a previous abnormal results of pap test or HPV DNA test. Genotyping, viral load and co-infections were determined. For each patient, the individual expression of 2549 microRNAs were evaluated by microarray and qPCR. (3) Results: Our data demonstrates that the microRNAs were commonly expressed in tissues biopsies. miR 4485-5p, miR4485-3p and miR-4497 were highly down-regulated in tissue biopsies with HPV precancerous cervical lesions. (4) Conclusions: the introduction of a microRNA analysis panel can improve early diagnosis, understand the nature of the lesion and, consequently, improve the clinical management of patients with HPV precancerous cervical lesions.

MLcps: machine learning cumulative performance score for classification problems

BACKGROUND

Assessing the performance of machine learning (ML) models requires careful consideration of the evaluation metrics used. It is often necessary to utilize multiple metrics to gain a comprehensive understanding of a trained model's performance, as each metric focuses on a specific aspect. However, comparing the scores of these individual metrics for each model to determine the best-performing model can be time-consuming and susceptible to subjective user preferences, potentially introducing bias.

RESULTS

We propose the Machine Learning Cumulative Performance Score (MLcps), a novel evaluation metric for classification problems. MLcps integrates several precomputed evaluation metrics into a unified score, enabling a comprehensive assessment of the trained model's strengths and weaknesses. We tested MLcps on 4 publicly available datasets, and the results demonstrate that MLcps provides a holistic evaluation of the model's robustness, ensuring a thorough understanding of its overall performance.

CONCLUSIONS

By utilizing MLcps, researchers and practitioners no longer need to individually examine and compare multiple metrics to identify the best-performing models. Instead, they can rely on a single MLcps value to assess the overall performance of their ML models. This streamlined evaluation process saves valuable time and effort, enhancing the efficiency of model evaluation. MLcps is available as a Python package at https://pypi.org/project/MLcps/.

IFN-I inducible miR-3614-5p targets ADAR1 isoforms and fine tunes innate immune activation

Regulation of innate immune responses is essential for maintenance of immune homeostasis and development of an appropriate immunity against microbial infection. We show here that miR-3614-5p, product of the TRIM25 host gene, is induced by type I interferon (IFN-I) in several human non-immune and immune cell types, in particular in primary myeloid cells. Studies in HeLa cells showed that miR-3614-5p represses both p110 and p150 ADAR1 and reduces constitutive and IFN-induced A-to-I RNA editing. In line with this, activation of innate sensors and expression of IFN-β and the pro-inflammatory IL-6 are promoted. MiR-3614-5p directly targets ADAR1 transcripts by binding to one specific site in the 3’UTR. Moreover, we could show that endogenous miR-3614-5p is associated with Ago2 and targets ADAR1 in IFN-stimulated cells. Overall, we propose that, by reducing ADAR1, IFN-I-induced miR-3614-5p contributes to lowering the activation threshold of innate sensors. Our findings provide new insights into the role of miR-3614-5p, placing it as a potential fine tuner of dsRNA metabolism, cell homeostasis and innate immunity.

A novel microfluidic RNA chip for direct, single-nucleotide specific, rapid and partially-degraded RNA detection

Direct RNA detection is critical for providing the RNA insights into gene expression profiling, noncoding RNAs, RNA-associated diseases and pathogens, without reverse transcription. However, classical RNA analysis usually requires RT-PCR, which can cause bias amplification and quantitation errors. To address this challenge, herein we report a microfluidic RNA chip (the microchip prototype) for direct RNA detection, which is primarily based on RNA extension and labeling with DNA polymerase. This detection strategy is of high specificity (discriminating against single-nucleotide differences), rapidity, accuracy, nuclease resistance, and reusability. Further, we have successfully detected disease-associated RNAs in clinical samples, demonstrating its great potentials in biomedical research and clinical diagnosis.

Comparison of different protocols of RNA preparation from circulating blood for RNA sequencing

OBJECTIVE

Circulating miRNAs have been extensively used in studies of neurological diseases. Thus, methods to extract high quantity total RNA for RNA sequencing (RNA-seq) and real-time quantitative polymerase chain reaction (RT-qPCR) are needed. However, the extraction of sufficient high-quality nucleic acids from circulating blood is difficult. Differences in eccentricity, cryopreservation conditions and extraction methods may affect RNA quantity and quality. Here, we systematically compared six blood-RNA extraction protocols (protocols 1, 2, 3, 4, 5, and 6; see the methods section for details).

RESULTS

Protocol 1 yielded the highest quality and quantity of RNA; protocol 2, protocol 5 and protocol 6 produced RNA of intermediate quality; and protocols 3 and 4 yielded the lowest quality RNA. The RNA integrity number (RIN) for isolated RNA was > 9.0 when protocol 1 or protocol 2 was used, > 8.0 when protocol 5 was used, and > 7.0 when protocol 6 was used; lower values were obtained when protocol 3 or 4 was used. The RNA extracted from circulating blood using protocol 1 was most intact and suitable for RT-qPCR and RNA-seq.

CONCLUSIONS

The quality of RNA extracted from circulating blood is affected by high-speed centrifugation and cryopreservation. Adding an RNA stabilizer during the cryopreservation of circulating blood significantly improved RNA quality and quantity. The quality of extracted RNA from circulating blood is improved when using TRIzol relative to that attained with a commercial kit.

Bayesian Modeling of Spatial Molecular Profiling Data via Gaussian Process

MOTIVATION

The location, timing, and abundance of gene expression (both mRNA and proteins) within a tissue define the molecular mechanisms of cell functions. Recent technology breakthroughs in spatial molecular profiling, including imaging-based technologies and sequencing-based technologies, have enabled the comprehensive molecular characterization of single cells while preserving their spatial and morphological contexts. This new bioinformatics scenario calls for effective and robust computational methods to identify genes with spatial patterns.

RESULTS

We represent a novel Bayesian hierarchical model to analyze spatial transcriptomics data, with several unique characteristics. It models the zero-inflated and over-dispersed counts by deploying a zero-inflated negative binomial model that greatly increases model stability and robustness. Besides, the Bayesian inference framework allows us to borrow strength in parameter estimation in a de novo fashion. As a result, the proposed model shows competitive performances in accuracy and robustness over existing methods in both simulation studies and two real data applications.

AVAILABILITY

The related R/C ++ source code is available at https://github.com/Minzhe/BOOST-GP.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Identification of potential plasma biomarkers in early-stage nasopharyngeal carcinoma-derived exosomes based on RNA sequencing

BACKGROUND

Early diagnosis of nasopharyngeal carcinoma (NPC) is vital to improve the prognosis of these patients. However, early diagnosis of NPC is typically challenging. Therefore, we explored the pathogenetic roles and associated mechanisms of exosomes in plasma of patients with early-stage NPC.

METHODS

Exosomes in plasma were extracted by ultra-high-speed centrifugation. Western blot and transmission electron microscopy (TEM) were used to verify the purity of exosomes. The sequencing data (6 plasma samples from healthy volunteers vs. 6 NPC plasma samples) were analyzed by principal component analysis (PCA), DESeq2, gene ontology (GO), Kyoto encyclopedia of genes and genomes (KEGG), and TargetScan. The differentially expressed miRNAs (DEmiRNAs) were obtained from the dataset (GSE118720) downloaded from the Gene Expression Omnibus (GEO) repository. Additionally, the datasets downloaded from the GEO database (GSE12452, GSE13597, GSE53819, GSE64634) were used to predict the target genes and functions of hsa-miR-1301-3p. qPCR was applied to verify the differences in the expressions of hsa-miR-1301-3p between 10 normal plasma and 10 NPC plasma samples.

RESULTS

Western blot, TEM, and Nanoparticle Tracking Analysis showed adequate purity of the extracted exosomes. RNA-seq analysis revealed 21 upregulated miRNAs, and 10 downregulated miRNAs in plasma exosomes of early-stage NPC patients. GO analysis showed that the target genes of DEmiRNAs were mainly enriched in DNA synthesis and transcription regulation. KEGG analysis revealed that DEmiRNAs were mainly enriched in PI3K-Akt and MAPK signaling pathways. Moreover, the expression of hsa-mir-1301-3p was verified to be significantly upregulated in enlarged samples of plasma exosomes.

CONCLUSIONS

We identified several DEmiRNAs extracted from tumor-derived exosomes between normal plasma and early-stage NPC plasma. Bioinformatics analyses indicated that these DEmiRNAs may be related to NPC development. Our study may provide novel insights into underlying biomarkers and mechanisms of plasma exosomes in early-stage NPC.

AGO-accessible anticancer siRNAs designed with synergistic miRNA-like activity

Small interfering RNAs (siRNAs) therapeutically induce RNA interference (RNAi) of disease-causing genes, but they also silence hundreds of seed-matched off-targets as behaving similar to microRNAs (miRNAs). miRNAs control the pathophysiology of tumors, wherein their accessible binding sites can be sequenced by Argonaute crosslinking immunoprecipitation (AGO CLIP). Herein, based on AGO CLIP, we develop potent anticancer siRNAs utilizing miRNA-like activity (mi/siRNAs). The mi/siRNAs contain seed sequences (positions 2-7) of tumor-suppressive miRNAs while maintaining perfect sequence complementarity to the AGO-accessible tumor target sites. Initially, host miRNA interactions with human papillomavirus 18 (HPV18) were identified in cervical cancer by AGO CLIP, revealing tumor-suppressive activity of miR-1/206 and miR-218. Based on the AGO-miRNA binding sites, mi/siRNAs were designed to target E6 and E7 (E6/E7) transcript with seed sequences of miR-1/206 (206/E7) and miR-218 (218/E7). Synergistic anticancer activity of 206/E7 and 218/E7 was functionally validated and confirmed via RNA sequencing and in vivo xenograft models (206/E7). Other mi/siRNA sequences were additionally designed for cervical, ovarian, and breast cancer, and available as an online tool (http://ago.korea.ac.kr/misiRNA); some of the mi/siRNAs were validated for their augmented anticancer activity (206/EphA2 and 206/Her2). mi/siRNAs could coordinate miRNA-like activity with robust siRNA function, demonstrating the potential of AGO CLIP analysis for RNAi therapeutics.

MicroRNA-3613-3p functions as a tumor suppressor and represents a novel therapeutic target in breast cancer

BACKGROUND

MicroRNAs have been reported to participate in tumorigenesis, treatment resistance, and tumor metastasis. Novel microRNAs need to be identified and investigated to guide the clinical prognosis or therapy for breast cancer.

METHOD

The copy number variations (CNVs) of MIR3613 from Cancer Genome Atlas (TCGA) or Cancer Cell Line Encyclopedia (CCLE) were analyzed, and its correlation with breast cancer subtypes or prognosis was investigated. The expression level of miR-3613-3p in tumor tissues or serum of breast cancer patients was detected using in situ hybridization and qPCR. Gain-of-function studies were performed to determine the regulatory role of miR-3613-3p on proliferation, apoptosis, and tumor sphere formation of human breast cancer cells MDA-MB-231 or MCF-7. The effects of miR-3613-3p on tumor growth or metastasis in an immunocompromised mouse model of MDA-MB-231-luciferase were explored by intratumor injection of miR-3613-3p analogue. The target genes, interactive lncRNAs, and related signaling pathways of miR-3613-3p were identified by bioinformatic prediction and 3'-UTR assays.

RESULTS

We found that MIR3613 was frequently deleted in breast cancer genome and its deletion was correlated with the molecular typing, and an unfavorable prognosis in estrogen receptor-positive patients. MiR-3613-3p level was also dramatically lower in tumor tissues or serum of breast cancer patients. Gain-of-function studies revealed that miR-3613-3p could suppress proliferation and sphere formation and promote apoptosis in vitro and impeded tumor growth and metastasis in vivo. Additionally, miR-3613-3p might regulate cell cycle by targeting SMS, PAFAH1B2, or PDK3 to restrain tumor progression.

CONCLUSION

Our findings indicate a suppressive role of miR-3613-3p in breast cancer progression, which may provide an innovative marker or treatment for breast cancer patients.

microRNA-seq of cartilage reveals an overabundance of miR-140-3p which contains functional isomiRs

miR-140 is selectively expressed in cartilage. Deletion of the entire Mir140 locus in mice results in growth retardation and early-onset osteoarthritis-like pathology; however, the relative contribution of miR-140-5p or miR-140-3p to the phenotype remains to be determined. An unbiased small RNA sequencing approach identified miR-140-3p as significantly more abundant (>10-fold) than miR-140-5p in human cartilage. Analysis of these data identified multiple miR-140-3p isomiRs differing from the miRBase annotation at both the 5' and 3' end, with >99% having one of two seed sequences (5' bases 2-8). Canonical (miR-140-3p.2) and shifted (miR-140-3p.1) seed isomiRs were overexpressed in chondrocytes and transcriptomics performed to identify targets. miR-140-3p.1 and miR-140-3p.2 significantly down-regulated 694 and 238 genes, respectively, of which only 162 genes were commonly down-regulated. IsomiR targets were validated using 3'UTR luciferase assays. miR-140-3p.1 targets were enriched within up-regulated genes in rib chondrocytes of Mir140-null mice and within down-regulated genes during human chondrogenesis. Finally, through imputing the expression of miR-140 from the expression of the host gene WWP2 in 124 previously published data sets, an inverse correlation with miR-140-3p.1 predicted targets was identified. Together these data suggest the novel seed containing isomiR miR-140-3p.1 is more functional than original consensus miR-140-3p seed containing isomiR.

Upregulation of miR-205 induces CHN1 expression, which is associated with the aggressive behaviour of cervical cancer cells and correlated with lymph node metastasis

BACKGROUND

Cervical cancer is the leading cause of cancer-related death in women worldwide. However, the mechanisms mediating the development and progression of cervical cancer are unclear. In this study, we aimed to elucidate the roles of microRNAs and a1-chimaerin (CHN1) protein in cervical cancer progression.

METHODS

The expression of miR-205 and CHN1 protein was investigated by in situ hybridisation and immunohistochemistry. We predicted the target genes of miR-205 using software prediction and dual luciferase assays. The expression of mRNAs and proteins was tested by qRT-PCR and western blotting respectively. The ability of cell growth, migration and invasion was evaluated by CCK-8 and transwell. Cell apoptosis was analysed by flow cytometry analysis.

RESULTS

We found that miR-205 and CHN1 were highly expressed in human cervical cancer tissue compared with paired normal cervical tissues. The CHN1 gene was shown to be targeted by miR-205 in HeLa cells. Interestingly, transfection with miR-205 mimic upregulated CHN1 mRNA and protein, while miR-205 inhibitor downregulated CHN1 in high-risk and human papilloma virus (HPV)-negative human cervical cancer cells in vitro,. These data suggested that miR-205 positively regulated the expression of CHN1. Furthermore, the miR-205 mimic promoted cell growth, apoptosis, migration, and invasion in high-risk and HPV-negative cervical cancer cells, while the miR-205 inhibitor blocked these biological processes. Knockdown of CHN1 obviously reduced the aggressive cellular behaviours induced by upregulation of miR-205, suggesting that miR-205 positively regulated CHN1 to mediate these cell behaviours during the development of cervical cancer. Furthermore, CHN1 was correlated with lymph node metastasis in clinical specimens.

CONCLUSIONS

Our findings showed that miR-205 positively regulated CHN1 to mediate cell growth, apoptosis, migration, and invasion during cervical cancer development, particularly for high-risk HPV-type cervical cancer. These findings suggested that dysregulation of miR-205 and subsequent abnormalities in CHN1 expression promoted the oncogenic potential of human cervical cancer.

mirnaQC: a webserver for comparative quality control of miRNA-seq data

Although miRNA-seq is extensively used in many different fields, its quality control is frequently restricted to a PhredScore-based filter. Other important quality related aspects like microRNA yield, the fraction of putative degradation products (such as rRNA fragments) or the percentage of adapter-dimers are hard to assess using absolute thresholds. Here we present mirnaQC, a webserver that relies on 34 quality parameters to assist in miRNA-seq quality control. To improve their interpretability, quality attributes are ranked using a reference distribution obtained from over 36 000 publicly available miRNA-seq datasets. Accepted input formats include FASTQ and SRA accessions. The results page contains several sections that deal with putative technical artefacts related to library preparation, sequencing, contamination or yield. Different visualisations, including PCA and heatmaps, are available to help users identify underlying issues. Finally, we show the usefulness of this approach by analysing two publicly available datasets and discussing the different quality issues that can be detected using mirnaQC.

MiR-944/CISH mediated inflammation via STAT3 is involved in oral cancer malignance by cigarette smoking

•

Down-regulation of CISH in OSCC tissues and cell lines.

•

CISH mediates cellular functions through STAT3 inhibition.

•

MiR-944 regulates cellular functions through direct binding of CISH.

•

Cigarette smoking-mediated miR-944/CISH/STAT3 axis plays a role in oral carcinogenesis.

The cytokine-inducible Src homology 2-containing protein (CISH) is an endogenous suppressors of signal transduction and activator of transcription (STAT) and acts as a key negative regulator of inflammatory cytokine responses. Downregulation of CISH has been reported to associate with increased activation of STAT and enhanced inflammatory pathways. However, whether microRNAs (miRNAs) play a crucial role in CISH/STAT regulation in oral squamous cell carcinoma (OSCC) remains unknown. The expression of CISH on OSCC patients was determine by quantitative real-time PCR (qRT-PCR) and immunohistochemistry. Specific targeting by miRNAs was determined by software prediction, luciferase reporter assay, and correlation with target protein expression. The functions of miR-944 and CISH were accessed by transwell migration and invasion analyses using gain- and loss-of-function approaches. Enzyme-linked immunosorbent assay (ELISA) and qRT-PCR were used to evaluate the pro-inflammation cytokines expression under the miR-944, CISH, NNK or combinations treatment. We found that the CISH protein, which modulates STAT3 activity, as a direct target of miR-944. CISH protein was significantly down-regulated in OSCC patients and cell lines and its level was inversely correlated with miR-944 expression. The miR-944-induced STAT3 phosphorylation, pro-inflammation cytokines secretion, migration and invasion were abolished by CISH restoration, suggesting that the oncogenic activity of miR-944 is CISH dependent. Furthermore, tobacco extract (NNK) may contribute to miR-944 induction and STAT3 activation. Antagomir-mediated inactivation of miR-944 prevented the NNK-induced STAT3 phosphorylation and pro-inflammation cytokines secretion. Altogether, these data demonstrate that NNK-induced miR944 expression plays an important role in CISH/STAT3-mediated inflammatory response and activation of tumor malignancy.

GARS: Genetic Algorithm for the identification of a Robust Subset of features in high-dimensional datasets

BACKGROUND

Feature selection is a crucial step in machine learning analysis. Currently, many feature selection approaches do not ensure satisfying results, in terms of accuracy and computational time, when the amount of data is huge, such as in 'Omics' datasets.

RESULTS

Here, we propose an innovative implementation of a genetic algorithm, called GARS, for fast and accurate identification of informative features in multi-class and high-dimensional datasets. In all simulations, GARS outperformed two standard filter-based and two 'wrapper' and one embedded' selection methods, showing high classification accuracies in a reasonable computational time.

CONCLUSIONS

GARS proved to be a suitable tool for performing feature selection on high-dimensional data. Therefore, GARS could be adopted when standard feature selection approaches do not provide satisfactory results or when there is a huge amount of data to be analyzed.

qtQDA: quantile transformed quadratic discriminant analysis for high-dimensional RNA-seq data

Classification on the basis of gene expression data derived from RNA-seq promises to become an important part of modern medicine. We propose a new classification method based on a model where the data is marginally negative binomial but dependent, thereby incorporating the dependence known to be present between measurements from different genes. The method, called qtQDA, works by first performing a quantile transformation (qt) then applying Gaussian quadratic discriminant analysis (QDA) using regularized covariance matrix estimates. We show that qtQDA has excellent performance when applied to real data sets and has advantages over some existing approaches. An R package implementing the method is also available on https://github.com/goknurginer/qtQDA.

Human Papillomavirus Infections, Cervical Cancer and MicroRNAs: An Overview and Implications for Public Health

Human Papillomavirus (HPV) is among the most common sexually transmitted infections in both females and males across the world that generally do not cause symptoms and are characterized by high rates of clearance. Persistent infections due at least to twelve well-recognized High-Risk (HR) or oncogenic genotypes, although less frequent, can occur, leading to diseases and malignancies, principally cervical cancer. Three vaccination strategies are currently available for preventing certain HR HPVs-associated diseases, infections due to HPV6 and HPV11 low-risk types, as well as for providing cross-protection against non-vaccine genotypes. Nevertheless, the limited vaccine coverage hampers reducing the burden of HPV-related diseases globally. For HR HPV types, especially HPV16 and HPV18, the E6 and E7 oncoproteins are needed for cancer development. As for other tumors, even in cervical cancer, non-coding microRNAs (miRNAs) are involved in posttranscriptional regulation, resulting in aberrant expression profiles. In this study, we provide a summary of the epidemiological background for HPV occurrence and available immunization programs. In addition, we present an overview of the most relevant evidence of miRNAs deregulation in cervical cancer, underlining that targeting these biomolecules could lead to wide translational perspectives, allowing better diagnosis, prognosis and therapeutics, and with valuable applications in the field of prevention. The literature on this topic is rapidly growing, but advanced investigations are required to achieve more consistent findings on the up-regulated and down-regulated miRNAs in cervical carcinogenesis. Because the expression of miRNAs is heterogeneously reported, it may be valuable to assess factors and risks related to individual susceptibility.

An insight of microRNAs performance in carcinogenesis and tumorigenesis; an overview of cancer therapy

Importance of dysregulation and expression of microRNAs (miRNAs) has been confiemed in many disorders comprising cancer. In this way, different approaches to induce reprogramming from one cell type to another in oerder to control the cell normal mechanisem, comprising microRNAs, combinatorial small molecules, exosome-mediated reprogramming, embryonic microenvironment and also lineage-specific transcription agents, are involved in cell situation. Meaningly, besides the above factors, microRNAs are so special and have an impressive role in cell reprogramming. One of the main applications of cancer cell reprogramming is it's ability in therapeutic approach. Many insights in reprogramming mechanism have been recommended, and determining improvment has been aknolwged to develop reprogramming efficiency and possibility, permiting it to appear as practical therapy against all cancers. Conspiciously, the recent studies on the fluctuations and performance of microRNAs,small endogenous non-coding RNAs, as notable factors in carcinogenesis and tumorigenesis, therapy resistance and metastasis and as new non-invasive cancer biomarkers has a remarkable attention. This is due to their unique dysregulated signatures throughout tumor progression. Recognising miRNAs signatures capable of anticipating therapy response and metastatic onset in cancers might enhance diagnosis and therapy. According to the growing reports on miRNAs as novel non-invasive biomarkers in various cancers as a main regulators of cancers drug resistance or metastasis, the quest on whether some miRNAs have the ability to regulate both simultaneously is inevitable, yet understudied. The combination of genetic diagnosis using next generation sequencing and targeted therapy may contribute to the effective precision medicine for cancer therapy. Here, we want to review the practical application of microRNAs performance in carcinogenesis and tumorigenesis in cancer therapy.

MLSeq: Machine learning interface for RNA-sequencing data

BACKGROUND AND OBJECTIVE

In the last decade, RNA-sequencing technology has become method-of-choice and prefered to microarray technology for gene expression based classification and differential expression analysis since it produces less noisy data. Although there are many algorithms proposed for microarray data, the number of available algorithms and programs are limited for classification of RNA-sequencing data. For this reason, we developed MLSeq, to bring not only frequently used classification algorithms but also novel approaches together and make them available to be used for classification of RNA sequencing data. This package is developed using R language environment and distributed through BIOCONDUCTOR network.

METHODS

Classification of RNA-sequencing data is not straightforward since raw data should be preprocessed before downstream analysis. With MLSeq package, researchers can easily preprocess (normalization, filtering, transformation etc.) and classify raw RNA-sequencing data using two strategies: (i) to perform algorithms which are directly proposed for RNA-sequencing data structure or (ii) to transform RNA-sequencing data in order to bring it distributionally closer to microarray data structure, and perform algorithms which are developed for microarray data. Moreover, we proposed novel algorithms such as voom (an acronym for variance modelling at observational level) based nearest shrunken centroids (voomNSC), diagonal linear discriminant analysis (voomDLDA), etc. through MLSeq.

MATERIALS

Three real RNA-sequencing datasets (i.e cervical cancer, lung cancer and aging datasets) were used to evalute model performances. Poisson linear discriminant analysis (PLDA) and negative binomial linear discriminant analysis (NBLDA) were selected as algorithms based on dicrete distributions, and voomNSC, nearest shrunken centroids (NSC) and support vector machines (SVM) were selected as algorithms based on continuous distributions for model comparisons. Each algorithm is compared using classification accuracies and sparsities on an independent test set.

RESULTS

The algorithms which are based on discrete distributions performed better in cervical cancer and aging data with accuracies above 0.92. In lung cancer data, the most of algorithms performed similar with accuracies of 0.88 except that SVM achieved 0.94 of accuracy. Our voomNSC algorithm was the most sparse algorithm, and able to select 2.2% and 6.6% of all features for cervical cancer and lung cancer datasets respectively. However, in aging data, sparse classifiers were not able to select an optimal subset of all features.

CONCLUSION

MLSeq is comprehensive and easy-to-use interface for classification of gene expression data. It allows researchers perform both preprocessing and classification tasks through single platform. With this property, MLSeq can be considered as a pipeline for the classification of RNA-sequencing data.

miR-3622a promotes proliferation and invasion of bladder cancer cells by downregulating LASS2

As a tumor metastasis suppressor gene, LASS2 has been found to be negatively associated with the stage of bladder cancer and overall survival of patients. However, the mechanisms regulating LASS2 in bladder cancer remain poorly understood. Here, we aim to identify a miRNA that targets LASS2 from bladder cancer-associated miRNAs and to reveal its potential functions in bladder cancer cells. Through miRNA microarray and bioinformatics analyses, we identified miR-3622a as a negative regulator of LASS2. The expression levels of miR-3622a in bladder cancer tissues were negatively correlated with the overall survival of patients. Overexpression of miR-3622a significantly increased the proliferation and invasion abilities of bladder cancer cells. In conclusion, our results indicate that miR-3622a promotes the proliferation and invasion of bladder cancer cells by downregulating LASS2.

Model-free feature screening for categorical outcomes: Nonlinear effect detection and false discovery rate control

Feature screening has become a real prerequisite for the analysis of high-dimensional genomic data, as it is effective in reducing dimensionality and removing redundant features. However, existing methods for feature screening have been mostly relying on the assumptions of linear effects and independence (or weak dependence) between features, which might be inappropriate in real practice. In this paper, we consider the problem of selecting continuous features for a categorical outcome from high-dimensional data. We propose a powerful statistical procedure that consists of two steps, a nonparametric significance test based on edge count and a multiple testing procedure with dependence adjustment for false discovery rate control. The new method presents two novelties. First, the edge-count test directly targets distributional difference between groups, therefore it is sensitive to nonlinear effects. Second, we relax the independence assumption and adapt Efron's procedure to adjust for the dependence between features. The performance of the proposed procedure, in terms of statistical power and false discovery rate, is illustrated by simulated data. We apply the new method to three genomic datasets to identify genes associated with colon, cervical and prostate cancers.

Classifying next-generation sequencing data using a zero-inflated Poisson model

Motivation

With the development of high-throughput techniques, RNA-sequencing (RNA-seq) is becoming increasingly popular as an alternative for gene expression analysis, such as RNAs profiling and classification. Identifying which type of diseases a new patient belongs to with RNA-seq data has been recognized as a vital problem in medical research. As RNA-seq data are discrete, statistical methods developed for classifying microarray data cannot be readily applied for RNA-seq data classification. Witten proposed a Poisson linear discriminant analysis (PLDA) to classify the RNA-seq data in 2011. Note, however, that the count datasets are frequently characterized by excess zeros in real RNA-seq or microRNA sequence data (i.e. when the sequence depth is not enough or small RNAs with the length of 18-30 nucleotides). Therefore, it is desired to develop a new model to analyze RNA-seq data with an excess of zeros.

Results

In this paper, we propose a Zero-Inflated Poisson Logistic Discriminant Analysis (ZIPLDA) for RNA-seq data with an excess of zeros. The new method assumes that the data are from a mixture of two distributions: one is a point mass at zero, and the other follows a Poisson distribution. We then consider a logistic relation between the probability of observing zeros and the mean of the genes and the sequencing depth in the model. Simulation studies show that the proposed method performs better than, or at least as well as, the existing methods in a wide range of settings. Two real datasets including a breast cancer RNA-seq dataset and a microRNA-seq dataset are also analyzed, and they coincide with the simulation results that our proposed method outperforms the existing competitors.

Availability and implementation

The software is available at http://www.math.hkbu.edu.hk/∼tongt.

Contact

xwan@comp.hkbu.edu.hk or tongt@hkbu.edu.hk.

Supplementary information

Supplementary data are available at Bioinformatics online.

Small RNA sequencing of sessile serrated polyps identifies microRNA profile associated with colon cancer

Sessile serrated adenoma/polyps (SSA/Ps) of the colon account for 20-30% of all colon cancers. Small non-coding RNAs, including microRNAs (miRNAs), may function as oncogenes or tumor suppressor genes involved in cancer development. Small RNA sequencing (RNA-seq) was used to characterize miRNA profiles in SSA/Ps, hyperplastic polyps (HPs), adenomatous polyps and paired uninvolved colon. Our 108 small RNA-seq samples' results were compared to small RNA-seq data from 212 colon cancers from the Cancer Genome Atlas. Twenty-three and six miRNAs were differentially expressed in SSA/Ps compared to paired uninvolved colon and HPs, respectively. Differential expression of MIR31-5p, MIR135B-5p and MIR378A-5p was confirmed by RT-qPCR. SSA/P-specific miRNAs are similarly expressed in colon cancers containing genomic aberrations described in serrated cancers. Correlation of miRNA expression with consensus molecular subtypes suggests more than one subtype is associated with the serrated neoplasia pathway. Canonical pathway analysis suggests many of these miRNAs target growth factor signaling pathways.

Oncogenic microRNA signature for early diagnosis of cervical intraepithelial neoplasia and cancer

Cervical cancer is one of the leading causes of cancer death in women globally, despite the widespread use of cytology/human papillomavirus (HPV) screening. In the present study, we aimed to identify the potential role of microRNA (miRNA) as a diagnostic biomarker in the detection of cervical pre‐malignant lesions and cancer. In total, we recruited 582 patients with cervical diseases and 145 control individuals. The expression levels of six miRNAs (miR‐20a, miR‐92a, miR‐141, miR‐183*, miR‐210 and miR‐944) were found to be significantly up‐regulated in cervical cancer and pre‐malignant lesions compared to normal cervical samples, indicating that they are oncogenic miRNAs. Receiver operating characteristic curve analysis showed that these six miRNAs can be used to distinguish patients with cervical pre‐malignant lesions or cancer from normal individuals and they also had a good predictive performance, particularly in cervical lesions. Combined use of these six miRNAs further enhanced the diagnostic accuracy over any single miRNA marker, with an area under the curve of 0.998, 0.996 and 0.959, a diagnostic sensitivity of 97.9%, 97.2% and 91.4%, and a specificity of 98.6%, 96.6% and 87.6% for low‐grade lesions, high‐grade lesions and cancer, respectively. This six oncogenic miRNA signature may be suitable for use as diagnostic marker for cervical pre‐malignant lesions and cancer in the near future.

MiR-875 and miR-3144 switch the human papillomavirus 16 E6/E6* mRNA ratio through the EGFR pathway and a direct targeting effect

By employing bioinformatics scanning approaches and luciferase reporter, our previous study showed that two less common human miRNAs, miR-875 and miR-3144, target a conserved site in the genomes of most high-risk human papillomaviruses (HR-HPVs). In this study, we found that the site targeted by miR-875 and miR-3144 overlapped with the 5' alternative splice site of E6E7 transcripts in HPV16. Using HPV16+ SiHa cells, we showed that high levels of miR-875 and miR-3144 reduced the abundance of unspliced E6, while they promoted three E6* spliced transcripts and decreased the expression levels of E6/E7 oncoproteins and epidermal growth factor receptor (EGFR). A potential miR-875 target site was predicted in EGFR. Meanwhile, depletion of EGFR resulted in a failure to promote E6* but maintained the suppression of unspliced E6 driven by miR-875 and miR-3144. The data suggest that these two miRNAs switch the E6/E6* ratio through both the EGFR pathway and direct targeting. Here, we demonstrate for the first time that human miRNAs regulate the HPV splice isoforms. Furthermore, miRNA-875 and miRNA-3144 are only found in vertebrates and Homo sapiens, and the binding site in EGFR is highly conserved in Boreoeutheria. Our findings highlight the tumour-suppressing effect of miRNAs that possibly appeared in the late stage of biological evolution.

ABBREVIATIONS

A Bayesian hierarchical model for analyzing methylated RNA immunoprecipitation sequencing data

Background

The recently emerged technology of methylated RNA immunoprecipitation sequencing (MeRIP-seq) sheds light on the study of RNA epigenetics. This new bioinformatics question calls for effective and robust peaking calling algorithms to detect mRNA methylation sites from MeRIP-seq data.

Methods

We propose a Bayesian hierarchical model to detect methylation sites from MeRIP-seq data. Our modeling approach includes several important characteristics. First, it models the zero-inflated and over-dispersed counts by deploying a zero-inflated negative binomial model. Second, it incorporates a hidden Markov model (HMM) to account for the spatial dependency of neighboring read enrichment. Third, our Bayesian inference allows the proposed model to borrow strength in parameter estimation, which greatly improves the model stability when dealing with MeRIP-seq data with a small number of replicates. We use Markov chain Monte Carlo (MCMC) algorithms to simultaneously infer the model parameters in a de novo fashion. The R Shiny demo is available at https://qiwei.shinyapps.io/BaySeqPeak and the R/C ++ code is available at https://github.com/liqiwei2000/BaySeqPeak.

Results

In simulation studies, the proposed method outperformed the competing methods exomePeak and MeTPeak, especially when an excess of zeros were present in the data. In real MeRIP-seq data analysis, the proposed method identified methylation sites that were more consistent with biological knowledge, and had better spatial resolution compared to the other methods.

Conclusions

In this study, we develop a Bayesian hierarchical model to identify methylation peaks in MeRIP-seq data. The proposed method has a competitive edge over existing methods in terms of accuracy, robustness and spatial resolution.

MicroRNA-focused CRISPR-Cas9 library screen reveals fitness-associated miRNAs

MicroRNAs (miRNAs) are post-transcriptional gene regulators that play important roles in the control of cell fitness, differentiation, and development. The CRISPR-Cas9 gene-editing system is composed of the Cas9 nuclease in complex with a single guide RNA (sgRNA) and directs DNA cleavage at a predetermined site. Several CRISPR-Cas9 libraries have been constructed for genome-scale knockout screens of protein function; however, few libraries have included miRNA genes. Here we constructed a miRNA-focused CRISPR-Cas9 library that targets 1594 (85%) annotated human miRNA stem-loops. The sgRNAs in our LX-miR library are designed to have high on-target and low off-target activity, and each miRNA is targeted by four to five sgRNAs. We used this sgRNA library to screen for miRNAs that affect cell fitness of HeLa or NCI-N87 cells by monitoring the change in frequency of each sgRNA over time. By considering the expression in the tested cells and the dysregulation of the miRNAs in cancer specimens, we identified five HeLa pro-fitness and cervical cancer up-regulated miRNAs (miR-31-5p, miR-92b-3p, miR-146b-5p, miR-151a-3p, and miR-194-5p). Similarly, we identified six NCI-N87 pro-fitness and gastric cancer up-regulated miRNAs (miR-95-3p, miR-181a-5p, miR-188-5p, miR-196b-5p, miR-584-5p, and miR-1304-3p), as well as three anti-fitness and down-regulated miRNAs (let-7a-3p, miR-100-5p, and miR-149-5p). Some of those miRNAs are known to be oncogenic or tumor-suppressive, but others are novel. Taken together, the LX-miR library is useful for genome-wide unbiased screening to identify miRNAs important for cellular fitness and likely to be useful for other functional screens.

Novel tumor suppressor microRNA at frequently deleted chromosomal region 8p21 regulates epidermal growth factor receptor in prostate cancer

Genomic loss of chromosome (chr) 8p21 region, containing prostate-specific NKX3.1 gene, is a frequent alteration of the prostate cancer (PCa) oncogenome. We propose a novel, paradigm shifting hypothesis that this frequently deleted locus is also associated with a cluster of microRNA genes- miR-3622a/b- that are lost in PCa and play an important mechanistic role in progression and metastasis. In this study, we demonstrate the role of miR-3622b in prostate cancer. Expression analyses in a cohort of PCa clinical specimens and cell lines show that miR-3622b expression is frequently lost in prostate cancer. Low miR-3622b expression was found to be associated with tumor progression and poor biochemical recurrence-free survival. Further, our analyses suggest that miR-3622b expression is a promising prostate cancer diagnostic biomarker that exhibits 100% specificity and 66% sensitivity. Restoration of miR-3622b expression in PCa cell lines led to reduced cellular viability, proliferation, invasiveness, migration and increased apoptosis. miR-3622b overexpression in vivo induced regression of established prostate tumor xenografts pointing to its therapeutic potential. Further, we found that miR-3622b directly represses Epidermal Growth Factor Receptor (EGFR). In conclusion, our study suggests that miR-3622b plays a tumor suppressive role and is frequently downregulated in prostate cancer, leading to EGFR upregulation. Importantly, miR-3622b has associated diagnostic, prognostic and therapeutic potential. Considering the association of chr8p21 loss with poor prognosis, our findings are highly significant and support a novel concept that associates a long standing observation of frequent loss of a chromosomal region with a novel miRNA in prostate cancer.

Detection and Verification of Mammalian Mirtrons by Northern Blotting

microRNAs (miRNAs) have vital roles in regulating gene expression-contributing to major diseases like cancer and heart disease. Over the last decade, thousands of miRNAs have been discovered through high throughput sequencing-based annotation. Different classes have been described, as well as a great dynamic range of expression levels. While sequencing approaches provide insight into biogenesis and allow confident identification, there is a need for additional methods for validation and characterization. Northern blotting was one of the first techniques used for studying miRNAs, and remains one of the most valuable as it avoids enzymatic manipulation of miRNA transcripts. Blotting can also provide insight into biogenesis by revealing RNA processing intermediates. Compared to sequencing, however, northern blotting is a relatively insensitive technology. This creates a challenge for detecting low expressed miRNAs, particularly those produced by inefficient, non-canonical pathways. In this chapter, we describe a strategy to study such miRNAs by northern blotting that involves ectopic expression of both miRNAs and miRNA-binding Argonaute (Ago) proteins. Through use of epitope tags, this strategy also provides a convenient method for verification of small RNA competency to be loaded into regulatory complexes.

MicroRNA expression profiles in non‑epithelial ovarian tumors

Ovarian germ cell tumors (OGCTs) and sex cord stromal tumors (SCSTs) are rare gynecologic tumors that are derived from germ and stromal cells, respectively. Unlike their epithelial counterparts, molecular pathogenesis of these tumor types is still poorly understood. Here, we characterized microRNA (miRNA) expression profiles of 9 OGCTs (2 malignant and 7 benign) and 3 SCSTs using small RNA sequencing. We observed significant miRNA expression variations among the three tumor groups. To further demonstrate the biological relevance of our findings, we selected 12 miRNAs for validation in an extended cohort of 16 OGCTs (9 benign and 7 malignant) and 7 SCSTs by reverse transcription-quantitative polymerase chain reaction. Higher expression of miR‑373‑3p, miR‑372‑3p and miR‑302c‑3p and lower expression of miR‑199a‑5p, miR‑214‑5p and miR‑202‑3p were reproducibly observed in malignant OGCTs as compared to benign OGCTs or SCSTs. Comparing with benign OGCTs, miR‑202c‑3p and miR‑513c‑5p were more abundant in SCSTs. Additionally, we examined Beclin 1 (BECN1), a target of miR‑199a‑5p, in the clinical samples using western blot analysis. Our results show that BECN1 expression was higher in malignant OGCTs than benign OGCTs, which is concordant with their lower miR‑199a‑5p expression. This study suggests that these miRNAs may have potential value as tumor markers and implications for further understanding the molecular basis of these tumor types.

High-throughput approaches for screening and analysis of cell behaviors

The rapid development of new biomaterials and techniques to modify them challenge our capability to characterize them using conventional methods. In response, numerous high-throughput (HT) strategies are being developed to analyze biomaterials and their interactions with cells using combinatorial approaches. Moreover, these systematic analyses have the power to uncover effects of delivered soluble bioactive molecules on cell responses. In this review, we describe the recent developments in HT approaches that help identify cellular microenvironments affecting cell behaviors and highlight HT screening of biochemical libraries for gene delivery, drug discovery, and toxicological studies. We also discuss HT techniques for the analyses of cell secreted biomolecules and provide perspectives on the future utility of HT approaches in biomedical engineering.

voomDDA: discovery of diagnostic biomarkers and classification of RNA-seq data

RNA-Seq is a recent and efficient technique that uses the capabilities of next-generation sequencing technology for characterizing and quantifying transcriptomes. One important task using gene-expression data is to identify a small subset of genes that can be used to build diagnostic classifiers particularly for cancer diseases. Microarray based classifiers are not directly applicable to RNA-Seq data due to its discrete nature. Overdispersion is another problem that requires careful modeling of mean and variance relationship of the RNA-Seq data. In this study, we present voomDDA classifiers: variance modeling at the observational level (voom) extensions of the nearest shrunken centroids (NSC) and the diagonal discriminant classifiers. VoomNSC is one of these classifiers and brings voom and NSC approaches together for the purpose of gene-expression based classification. For this purpose, we propose weighted statistics and put these weighted statistics into the NSC algorithm. The VoomNSC is a sparse classifier that models the mean-variance relationship using the voom method and incorporates voom’s precision weights into the NSC classifier via weighted statistics. A comprehensive simulation study was designed and four real datasets are used for performance assessment. The overall results indicate that voomNSC performs as the sparsest classifier. It also provides the most accurate results together with power-transformed Poisson linear discriminant analysis, rlog transformed support vector machines and random forests algorithms. In addition to prediction purposes, the voomNSC classifier can be used to identify the potential diagnostic biomarkers for a condition of interest. Through this work, statistical learning methods proposed for microarrays can be reused for RNA-Seq data. An interactive web application is freely available at http://www.biosoft.hacettepe.edu.tr/voomDDA/.

miR-708-5p: a microRNA with emerging roles in cancer

MicroRNAs (miRNAs) are small non-coding RNAs that negatively regulate gene expression post-transcriptionally. They are crucial for normal development and maintaining homeostasis. Researchers have discovered that dysregulated miRNA expression contributes to many pathological conditions, including cancer. miRNAs can augment or suppress tumorigenesis based on their expression and transcribed targetome in various cell types. In recent years, researchers have begun to identify miRNAs commonly dysregulated in cancer. One recently identified miRNA, miR-708-5p, has been shown to have profound roles in promoting or suppressing oncogenesis in a myriad of solid and hematological tumors. This review highlights the diverse, sometimes controversial findings reported for miR-708-5p in cancer, and the importance of further exploring this exciting miRNA.

Analysis of the miR-34 family functions in breast cancer reveals annotation error of miR-34b

The microRNAs in the miR-34 family, consisting of miR-34a, miR-34b and miR-34c, are tumour suppressors. The annotated human miR-34b-5p has one additional base at the 5' end of the common miR-34 family seed sequence, compared to miR-34a-5p and miR-34c-5p. This extra base results in a shift of the seed sequence, which would affect the target gene repertoire and have functional consequences. During our studies of miR-34 functions, we investigated the precise sequence of mature miR-34b-5p in human cells by deep sequencing. We found that a miR-34b-5p without the extra base was the predominant form in both non-malignant and malignant cells derived from several human tissues, indicating that the miR-34b annotation is misleading. We evaluated the functional implications of the seed shift, by comparing the effect of mimics representing the alternative miR-34b-5p sequences in MDA-MB-231 cells. In contrast to the annotated miR-34b, the endogenously expressed miR-34b displayed tumour suppressive characteristics in vitro similarly to miR-34c. These data demonstrate the importance of determining the precise sequence of a mature microRNA before exploring miRNA functions.

A comprehensive simulation study on classification of RNA-Seq data

RNA sequencing (RNA-Seq) is a powerful technique for the gene-expression profiling of organisms that uses the capabilities of next-generation sequencing technologies. Developing gene-expression-based classification algorithms is an emerging powerful method for diagnosis, disease classification and monitoring at molecular level, as well as providing potential markers of diseases. Most of the statistical methods proposed for the classification of gene-expression data are either based on a continuous scale (eg. microarray data) or require a normal distribution assumption. Hence, these methods cannot be directly applied to RNA-Seq data since they violate both data structure and distributional assumptions. However, it is possible to apply these algorithms with appropriate modifications to RNA-Seq data. One way is to develop count-based classifiers, such as Poisson linear discriminant analysis and negative binomial linear discriminant analysis. Another way is to bring the data closer to microarrays and apply microarray-based classifiers. In this study, we compared several classifiers including PLDA with and without power transformation, NBLDA, single SVM, bagging SVM (bagSVM), classification and regression trees (CART), and random forests (RF). We also examined the effect of several parameters such as overdispersion, sample size, number of genes, number of classes, differential-expression rate, and the transformation method on model performances. A comprehensive simulation study is conducted and the results are compared with the results of two miRNA and two mRNA experimental datasets. The results revealed that increasing the sample size, differential-expression rate and decreasing the dispersion parameter and number of groups lead to an increase in classification accuracy. Similar with differential-expression studies, the classification of RNA-Seq data requires careful attention when handling data overdispersion. We conclude that, as a count-based classifier, the power transformed PLDA and, as a microarray-based classifier, vst or rlog transformed RF and SVM classifiers may be a good choice for classification. An R/BIOCONDUCTOR package, MLSeq, is freely available at https://www.bioconductor.org/packages/release/bioc/html/MLSeq.html.