analysis software bioarray software environment database  (Bioarray Inc)

Journal: Royal Society Open Science

Article Title: Concordance between RNA-sequencing data and DNA microarray data in transcriptome analysis of proliferative and quiescent fibroblasts

doi: 10.1098/rsos.150402

Figure Lengend Snippet: Differing reproducibility of microarray FC values. (Correlations between FC values (QUI/PRO) are shown for each pair of microarrays. The values in the upper diagonal contain the Pearson correlations, while those in the lower diagonal contain the Spearman correlations. Values not in parentheses represent correlations between untransformed FC values, while those in parentheses represent correlations between log-transformed FC values. As log transformation does not change the rank order, only one number is shown for the Spearman correlation for each pair. Correlations varied substantially depending on the pair of microarrays and the correlation metric used, ranging from −0.55 to 0.74.)

Article Snippet: Analysis of raw datasets was performed using the online microarray database software BioArray Software Environment (BASE) [ ], with which cross-channel correction and LOWESS normalization were performed.

Techniques: Microarray, Transformation Assay

Journal: Royal Society Open Science

Article Title: Concordance between RNA-sequencing data and DNA microarray data in transcriptome analysis of proliferative and quiescent fibroblasts

doi: 10.1098/rsos.150402

Figure Lengend Snippet: Differing reproducibility of microarray FC values. The log-transformed FC values from some pairs of microarrays were consistent with one another, while negative correlations were observed for other pairs. Panel ( a ) shows the relationship between the log-transformed FC values from microarray QP2 and those from microarray QP4, which exhibited a moderate to strong correlation ( r =0.74). By contrast, panel ( b ) shows the relationship between the log-transformed FC values from microarray QP1 and those from microarray QP4, which had a negative correlation ( r =−0.41).

Article Snippet: Analysis of raw datasets was performed using the online microarray database software BioArray Software Environment (BASE) [ ], with which cross-channel correction and LOWESS normalization were performed.

Techniques: Microarray, Transformation Assay

Journal: Royal Society Open Science

Article Title: Concordance between RNA-sequencing data and DNA microarray data in transcriptome analysis of proliferative and quiescent fibroblasts

doi: 10.1098/rsos.150402

Figure Lengend Snippet: High reproducibility of RNA-seq read counts, and moderate reproducibility of RNA-seq FC values. (The correlations between read counts (PRO1 versus PRO2 and QUI1 versus QUI2) and FC values ( QUI 1/ PRO 1 versus QUI 2/ PRO 2) are shown. Except for the Pearson correlations between non-log-transformed values, correlations between read counts were similar in magnitude to the correlations observed between microarray intensity values (electronic supplementary material, table S1). Correlations between FC values were close to those observed in the most highly correlated pairs of microarrays.)

Article Snippet: Analysis of raw datasets was performed using the online microarray database software BioArray Software Environment (BASE) [ ], with which cross-channel correction and LOWESS normalization were performed.

Techniques: Microarray

Journal: Royal Society Open Science

Article Title: Concordance between RNA-sequencing data and DNA microarray data in transcriptome analysis of proliferative and quiescent fibroblasts

doi: 10.1098/rsos.150402

Figure Lengend Snippet: Low concordance between RNA-seq data and DNA microarray data. (For each cell state (PRO and QUI), reads from the two RNA-seq replicates were pooled to give a single read count for each probe. Concordance was determined using both correlation between reads counts (for the RNA-seq data) and intensity values (for the microarray data), and between FC values (QUI/PRO). Correlations between read counts and intensity values were low, ranging from 0.18 to 0.41, as were correlations between FC values, which ranged from 0.02 to 0.23. ‘All’ represents the geometric mean of the FC values of the four microarrays. The correlations between the RNA-seq data and the mean of the four microarrays was better than between the RNA-seq data and any of the individual microarrays.)

Article Snippet: Analysis of raw datasets was performed using the online microarray database software BioArray Software Environment (BASE) [ ], with which cross-channel correction and LOWESS normalization were performed.

Techniques: Microarray

Journal: Royal Society Open Science

Article Title: Concordance between RNA-sequencing data and DNA microarray data in transcriptome analysis of proliferative and quiescent fibroblasts

doi: 10.1098/rsos.150402

Figure Lengend Snippet: Moderate concordance between the log-transformed RNA-seq FC values and the log-transformed geometric mean of the microarray FC values. The scatterplot shows that there was a moderate linear relationship between these two variables ( r =0.42).

Article Snippet: Analysis of raw datasets was performed using the online microarray database software BioArray Software Environment (BASE) [ ], with which cross-channel correction and LOWESS normalization were performed.

Techniques: Transformation Assay, RNA Sequencing, Microarray

Journal: Royal Society Open Science

Article Title: Concordance between RNA-sequencing data and DNA microarray data in transcriptome analysis of proliferative and quiescent fibroblasts

doi: 10.1098/rsos.150402

Figure Lengend Snippet: Moderate overlap between the probes with the highest FC values in the RNA-seq data and those with the highest FC values in the DNA microarray data. ( k represents the size of a given list (the 10, 50, 100, 500 or 1000 probes with the highest FC values), while n represents the number of probes in common between a list from the RNA-seq data and the corresponding list from the DNA microarray. The p -value represents the proportion of 10 000 random trials that had an equal or greater level of overlap than that actually observed. Thus, if none of the random trials had a greater level of overlap, then the p -value is 0. More overlapping probes than would be expected by chance were observed for all microarrays for k =100, 500 and 1000, while some arrays had statistically significant p -values for k =10 and k =50. ‘All’ represents the geometric mean of the FC values of the four microarrays.)

Article Snippet: Analysis of raw datasets was performed using the online microarray database software BioArray Software Environment (BASE) [ ], with which cross-channel correction and LOWESS normalization were performed.

Techniques: Microarray

Journal: Royal Society Open Science

Article Title: Concordance between RNA-sequencing data and DNA microarray data in transcriptome analysis of proliferative and quiescent fibroblasts

doi: 10.1098/rsos.150402

Figure Lengend Snippet: RNA-seq FC values correlate better with qRT-PCR FC values than do microarray FC values, although not to a statistically significant degree. Correlation coefficients are shown between the qRT-PCR FC values for 76 genes, and the FC values for corresponding probes in each individual microarray or in the combined RNA-seq replicates. ‘All’ represents the geometric mean of the FC values of the four microarrays. For all three correlation measures, the RNA-seq correlation was not significantly different ( p -value >0.05) from the correlation of any of the microarrays (Fisher's z -transformation).

Article Snippet: Analysis of raw datasets was performed using the online microarray database software BioArray Software Environment (BASE) [ ], with which cross-channel correction and LOWESS normalization were performed.

Techniques: Microarray, Transformation Assay