• 2019-10
  • 2019-11
  • 2020-03
  • 2020-07
  • 2020-08
  • 2021-03
  • br nCounter elements gene expression analysis Gene


    nCounter elements gene P 22077 analysis. Gene expression levels in training and testing sets were quantified via the nCounter Elements approach, according to manufacturer’s instructions (NanoString Technologies, Seattle, West Coast). Briefly, probes were designed to target sequences of interest and the corresponding Elements TagSet (consisting of a fluo-rescently labeled specific Reporter Tag and a biotiny-lated universal Capture Tag). The probes were hybridized with the Elements TagSet and 250 ng of RNA for a minimum of 16 hours at 67˚C in a thermo-cycler. Samples were stored at 4˚C for a maximum of 24 hours until purified in the nCounter Prep Station to remove unligated probes. Expression data was col-lected by direct digital counting of the target molecules in each sample using the nCounter Digital analyzer. Those samples with less than 10 counts were excluded from the analysis. Relative expression values for the genes analyzed were used to calculate the risk of pre-senting BC. If the predicted probability of the model was higher than the established cut-off point value, the samples were classified for the gene expression signa-ture as tumor sample. All the researchers from the Hos-pital Clinic involved in this analysis were blinded to the patients’ clinical data, ensuring the reliability of the results.
    NanoString raw data was processed in the R statistical environment (v3.3.2) and normaliza-tion was performed using the NanoStringNorm pack-age.15 The normalization setting was performed using the geometric mean of the 2 housekeeping genes
    Translational Research Montalbo et al 77
    Table I. Clinicopathological and demographic characteristics of the study population classified by (A) the study phase and
    (B) the participating center (testing set)
    (A) Discovery phase Training set Testing set Total Bladder cancer patients N (%) N (%) N (%) N (%)
    35 84 Urinary condition
    Testing set
    Hospital Clınic Fundacio Radboud University Virgen del University
    Puigvert Nijmegen Rocio of Vienna
    78 Montalbo et al Translational Research
    Table I. (Continued)
    Testing set
    Hospital Clınic Fundacio Radboud University Virgen del University
    Puigvert Nijmegen Rocio of Vienna
    Abbreviations: BPH, benign prostate history; MIBC, muscle invasive bladder cancer; NMIBC, non muscle-invasive bladder cancer; TURB, trans-urethral resection of the bladder.
    (BGUS, and PPIA). Logistic regression was used to generate diagnostic models. Performance was evalu-ated by Receiver Operating Characteristics (ROC) curves. Comparisons with a P value <0.05 were con-sidered statistically significant.
    ToppGene ( was used to detect functional enrichment of the 8 genes from the classifier. Gene-gene interaction networks for the genes of the model were built by the GeneMANIA Cytoscape 3.0.0 plugin.17 Physical, coexpression, and pathway gene-gene interactions were evaluated.
    Urinary biomarkers discovery. Initial exploratory assessment of the RNA-seq dataset was performed using principal component analysis. As shown, there is a noticeable overlap of urine samples from non-high-risk NMIBC and all other groups. Nevertheless, 521 genes were found to be specifically differentially expressed (P < 0.05) between non high-risk NMIBC and all the other groups (Fig 2 and Table S1).
    Classifier development. A total of 114 key differentially expressed genes selected from previous phase (false discovery rate 0.12 and absolute fold change 1.5) plus 16 selected genes from our previous classifiers7 9 (Table S2) were analyzed in 136 urine samples. There-after, 60 genes were selected (false discovery rate 0.12 and absolute fold change 1.5) to be tested in 79 independent additional samples (Table S3). Logistic regression analysis was used to generate an 8-gene expression classifier (ANXA10, IGF2, KIFC3, KRT20, LCN2,