A detailed analysis of next generation sequencing reads of microRNA expression in Barrett's Esophagus: absolute versus relative quantification

Citation metadata

From: BMC Research Notes(Vol. 7, Issue 1)
Publisher: BioMed Central Ltd.
Document Type: Article
Length: 2,914 words
Lexile Measure: 1300L

Document controls

Main content

Article Preview :

Author(s): In-Hee Lee1 , Xiaoman Hong2 , Sharad C Mathur3,4 , Mukut Sharma5,6 , Amit Rastogi7,8,9 , Prateek Sharma7,8,9 , Lane K Christenson2,8 and Ajay Bansal7,8,9


Next generation sequencing (NGS) is a significant advancement over hybridization-based microarrays for microRNA (miRNA) discovery. NGS can measure miRNA expression across several orders of magnitude from 1 to 10 7 . However, the quantitative interpretation of the primary output of NGS i.e. read counts for a miRNA sequence remains unclear. The current practice is to validate NGS findings by qRT-PCR[1-4]. However, the published studies have several limitations--a small number of biological samples[1, 2], primarily qualitative analysis[3], introduction of bias by selection for validation of only differentially expressed miRNA by NGS[3] and lack of guidance on low-versus high-abundance transcripts[1-4]. Specifically, several unanswered questions remain. How do NGS read counts correlate with Cq values on quantitative real-time polymerase chain reaction (qRT-PCR)? Is there a threshold copy number below which miRNA detection becomes unreliable? What is the overall sensitivity and specificity of NGS for identifying the miRNA of interest? How does NGS perform at absolute quantification of a transcript expression versus relative quantification between experimental and control groups? Does detection of differential expression of miRNA in a disease state depend on transcript abundance? Barrett's esophagus (BE) is a pre-malignant condition for rapidly increasing esophageal adenocarcinoma and is a complication of Gastroesophageal Reflux Disease (GERD)[5]. Here we present the systematic comparison of miRNA expression by NGS and qRT-PCR in well-characterized patients with BE and GERD.


Study design and patient selection

We previously sequenced the miRNA transcriptome in GERD and BE[6] and evaluated 14 differentially expressed miRNAs by qRT-PCR. For the current analysis, we analyzed an additional 22 miRNAs that were not differentially expressed by NGS. These additional miRNAs were randomly selected to represent the varying level of expression by NGS in GERD and BE tissues and to allow us to calculate NGS performance in an unbiased manner. Thus, we evaluated a total of 36 miRNA by qRT-PCR (Table 1). Patients with GERD and BE were selected from a prospective tissue and serum repository (Clinical Trials.gov # NCT00574327). The details of the repository, definitions and inclusion and exclusion criteria have been described previously[6]. The repository was created with approval by the Human Subjects Committee and the Research and Development Committee of the Institutional Review Board, Veterans Affairs Medical Center, Kansas City, Missouri. The repository has been annually approved since 2005. All patients sign an IRB approved informed consent prior to inclusion in the registry that allows us to store samples for future research related to GERD and BE. The approval number for the patient registry is ePROMISE PS0035 as determined under the institutional regulations. Briefly, BE is defined as presence of columnar lined esophagus on endoscopy with demonstration of intestinal metaplasia in biopsies. GERD is defined on the basis of presence of heartburn and/or regurgitation on a standardized and validated questionnaire. GERD patients are further sub-classified into those with erosive esophagitis (EE) and those without (Non-erosive...

Source Citation

Source Citation   

Gale Document Number: GALE|A540834634