Sensitive, reliable and robust circRNA detection from RNA-seq with CirComPara2

Gaffo, Enrico; Buratin, Alessia; Molin, Anna Dal; Bortoluzzi, Stefania

doi:10.1093/bib/bbab418

Cited by 37 publications

(18 citation statements)

References 70 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In contrast, each circRNA represents one single transcript, and the backsplice junction reads (BJRs) originate only from the speci c site of the circRNA sequence where the junction ends were joint [19] (Figure 1 a). Moreover, BJRs are computationally harder to identify than unspliced and linearly spliced reads as they require non-collinear alignments and additional processing to remove spurious hits [5], causing most circRNA detection tools to suffer from low detection rates [19,26].…”

Section: Circrna Expression Data Are Characterised By a High Proporti...mentioning

confidence: 99%

“…We veri ed this characteristic in 34 RNA-seq data sets of matched ribosomal RNA-depleted and circRNA-enriched libraries from 17 human tissues (Table 1). CirComPara2 [26] was used to obtain linear and circular read mappings on circRNAhost genes. We discriminated four read alignment sets representing the expression signal available for (i) estimating gene expression, (ii) studying alternative splicing, (iii) comparing the abundance of circular and linear transcripts expressed by a gene, and (iv) estimating circRNA abundance.…”

Section: Circrna Expression Data Are Characterised By a High Proporti...mentioning

confidence: 99%

“…However, their comparison was limited to parametric simulations based on a single real data set with a small number of replicates. Moreover, the parameter settings exploration was limited to the default for the two competitor methods, and only circRNAs expressed at moderate to high levels were considered, as is expected when selecting the circRNAs jointly predicted by multiple circRNA detection tools [25,26].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Systematic benchmarking of statistical methods to assess differential expression of circular RNAs

Buratin

Bortoluzzi

Gaffo

2023

Briefings in Bioinformatics

Self Cite

View full text Add to dashboard Cite

Circular RNAs (circRNAs) are covalently closed transcripts involved in critical regulatory axes, cancer pathways and disease mechanisms. CircRNA expression measured with RNA-seq has particular characteristics that might hamper the performance of standard biostatistical differential expression assessment methods (DEMs). We compared 38 DEM pipelines configured to fit circRNA expression data’s statistical properties, including bulk RNA-seq, single-cell RNA-seq (scRNA-seq) and metagenomics DEMs. The DEMs performed poorly on data sets of typical size. Widely used DEMs, such as DESeq2, edgeR and Limma-Voom, gave scarce results, unreliable predictions or even contravened the expected behaviour with some parameter configurations. Limma-Voom achieved the most consistent performance throughout different benchmark data sets and, as well as SAMseq, reasonably balanced false discovery rate (FDR) and recall rate. Interestingly, a few scRNA-seq DEMs obtained results comparable with the best-performing bulk RNA-seq tools. Almost all DEMs’ performance improved when increasing the number of replicates. CircRNA expression studies require careful design, choice of DEM and DEM configuration. This analysis can guide scientists in selecting the appropriate tools to investigate circRNA differential expression with RNA-seq experiments.

show abstract

Section: Circrna Expression Data Are Characterised By a High Proporti...mentioning

confidence: 99%

Section: Circrna Expression Data Are Characterised By a High Proporti...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Systematic benchmarking of statistical methods to assess differential expression of circular RNAs

Buratin

Bortoluzzi

Gaffo

2023

Briefings in Bioinformatics

Self Cite

View full text Add to dashboard Cite

show abstract

“…The three-prime exonuclease RNase R treatment and addition of poly-A tails are often used in the RNA-seq samples to identify circRNAs of low abundance [ 37 , 38 ]. To fully explore the circRNAs’ landscapes, increasing lines of evidence show that many computational pipelines are applied in the accurate annotation and quantification of circRNAs from RNA-seq data [ 39 , 40 , 41 ]. For instance, to achieve more comprehensive quantification, a new integrative approach named Short Read circRNA Pipeline (SRCP) was used to validate and quantify circRNAs with high sensitivity and a low number of false negatives [ 42 ].…”

Section: The Detection Technology Of Circrnasmentioning

confidence: 99%

Promising Roles of Circular RNAs as Biomarkers and Targets for Potential Diagnosis and Therapy of Tuberculosis

et al. 2022

View full text Add to dashboard Cite

Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb) infection, remains one of the most threatening infectious diseases worldwide. A series of challenges still exist for TB prevention, diagnosis and treatment, which therefore require more attempts to clarify the pathological and immunological mechanisms in the development and progression of TB. Circular RNAs (circRNAs) are a large class of non-coding RNA, mostly expressed in eukaryotic cells, which are generated by the spliceosome through the back-splicing of linear RNAs. Accumulating studies have identified that circRNAs are widely involved in a variety of physiological and pathological processes, acting as the sponges or decoys for microRNAs and proteins, scaffold platforms for proteins, modulators for transcription and special templates for translation. Due to the stable and widely spread characteristics of circRNAs, they are expected to serve as promising prognostic/diagnostic biomarkers and therapeutic targets for diseases. In this review, we briefly describe the biogenesis, classification, detection technology and functions of circRNAs, and, in particular, outline the dynamic, and sometimes aberrant changes of circRNAs in TB. Moreover, we further summarize the recent progress of research linking circRNAs to TB-related pathogenetic processes, as well as the potential roles of circRNAs as diagnostic biomarkers and miRNAs sponges in the case of Mtb infection, which is expected to enhance our understanding of TB and provide some novel ideas about how to overcome the challenges associated TB in the future.

show abstract

“…However, their comparison was limited to parametric simulations based on a single real data set with a small number of replicates. Moreover, the parameter settings exploration was limited to the default for the two competitor methods, and, nally, only circRNAs expressed at moderate to a high level were considered, as is expected when selecting the circRNAs jointly predicted by multiple circRNA detection tools [25,26].…”

Section: Introductionmentioning

confidence: 99%

Systematic benchmarking of bulk and single-cell RNA-seq statistical methods to assess differential expression of circular RNAs

Buratin

Bortoluzzi

Gaffo

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Background Circular RNAs (circRNAs) are transcripts with covalently joined 3' and 5' ends. They play critical regulatory roles, are involved in disease and cancer mechanisms, and are promising biomarkers. CircRNA differential abundance in high-throughput sequencing transcriptomics (RNA-seq) studies is commonly analysed using differential expression assessment methods (DEMs) devised for bulk RNA-seq. However, circRNA expression data from RNA-seq have a high proportion of small and zero counts that might hamper commonly used DEMs. Importantly, DEMs' performance has never been evaluated on circRNA data. Results We compared 38 DEM pipelines on circRNA expression data, exploring DEMs' parameter configurations conceived for small counts. We also evaluated single-cell RNA-seq (scRNA-seq) and metagenomics DEMs designed to handle sparse data. In general, the DEMs performed poorly on data sets of typical size. Widely used bulk RNA-seq DEMs, such as DESeq2, edgeR, and Limma-Voom, gave scarce results, unreliable predictions, or even contravened the expected behaviour with some parameter configurations. Limma-Voom achieved the most consistent performance throughout different benchmark data and, as well as SAMseq, reasonably balanced false discovery and recall rates. A few scRNA-seq DEMs obtained results comparable to the best performing bulk RNA-seq tools. Almost all DEMs' performance improved when increasing the number of replicates. Conclusions CircRNA expression data have particular characteristics distinguished from traditional bulk RNA-seq, requiring careful study design, choice of DEM, and DEM configuration. This study can guide scientists in selecting the appropriate tools for circRNA comparative RNA-seq analyses.

show abstract

Sensitive, reliable and robust circRNA detection from RNA-seq with CirComPara2

Cited by 37 publications

References 70 publications

Systematic benchmarking of statistical methods to assess differential expression of circular RNAs

Systematic benchmarking of statistical methods to assess differential expression of circular RNAs

Promising Roles of Circular RNAs as Biomarkers and Targets for Potential Diagnosis and Therapy of Tuberculosis

Systematic benchmarking of bulk and single-cell RNA-seq statistical methods to assess differential expression of circular RNAs

Contact Info

Product

Resources

About