Limits...
A comparison between droplet digital and quantitative PCR in the analysis of bacterial 16S load in lung tissue samples from control and COPD GOLD 2.

Sze MA, Abbasi M, Hogg JC, Sin DD - PLoS ONE (2014)

Bottom Line: We hypothesize that ddPCR is better at quantifying the total bacterial load in lung tissue versus qPCR.Total 16S counts were compared between the two methods.There was no difference in the average total 16S counts (P>0.05) between the two methods.

View Article: PubMed Central - PubMed

Affiliation: Centre for Heart Lung Innovation, St. Paul's Hospital, Departments of Medicine and Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC, Canada.

ABSTRACT

Background: Low biomass in the bacterial lung tissue microbiome utilizes quantitative PCR (qPCR) 16S bacterial assays at their limit of detection. New technology like droplet digital PCR (ddPCR) could allow for higher sensitivity and accuracy of quantification. These attributes are needed if specific bacteria within the bacterial lung tissue microbiome are to be evaluated as potential contributors to diseases such as chronic obstructive pulmonary disease (COPD). We hypothesize that ddPCR is better at quantifying the total bacterial load in lung tissue versus qPCR.

Methods: Control (n = 16) and COPD GOLD 2 (n = 16) tissue samples were obtained from patients who underwent lung resection surgery, were cut on a cryotome, and sections were assigned for use in quantitative histology or for DNA extraction. qPCR and ddPCR were performed on these samples using primers spanning the V2 region on the 16S rRNA gene along with negative controls. Total 16S counts were compared between the two methods. Both methods were assessed for correlations with quantitative histology measurements of the tissue.

Results: There was no difference in the average total 16S counts (P>0.05) between the two methods. However, the negative controls contained significantly lower counts in the ddPCR (0.55 ± 0.28 16S/uL) than in the qPCR assay (1.00 ± 0.70 16S copies) (P <0.05). The coefficient of variation was significantly lower for the ddPCR assay (0.18 ± 0.14) versus the qPCR assay (0.62 ± 0.29) (P<0.05).

Conclusion: Overall the ddPCR 16S assay performed better by reducing the background noise in 16S of the negative controls compared with 16S qPCR assay.

Show MeSH

Related in: MedlinePlus

Comparison of the Coefficients of Variation (CV) between QPCR and ddPCR.A) Direct comparison of the ddPCR CV against the QPCR CV. The ddPCR CV was significantly lower than those obtained using qPCR. B) A Bland-Altman plot of the ddPCR CV against the qPCR CV. On average there was a much larger CV for the qPCR than the ddPCR for each individual sample of 0.44 ± 0.29.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4199711&req=5

pone-0110351-g003: Comparison of the Coefficients of Variation (CV) between QPCR and ddPCR.A) Direct comparison of the ddPCR CV against the QPCR CV. The ddPCR CV was significantly lower than those obtained using qPCR. B) A Bland-Altman plot of the ddPCR CV against the qPCR CV. On average there was a much larger CV for the qPCR than the ddPCR for each individual sample of 0.44 ± 0.29.

Mentions: The ddPCR negative controls had a much smaller standard deviation versus the qPCR negative controls (0.28 versus 0.70). Both the qPCR and ddPCR detected a similar bacterial load for the control and moderate COPD groups. For the moderate COPD group, qPCR values were 2.32 ± 0.67 16S copies (mean ± SD) and ddPCR values were 2.80 ± 1.80 16S/uL and for the control group the qPCR and ddPCR values were 2.25 ± 1.55 16S copies and 2.36 ± 1.95 16S/uL (mean ± SD) respectively. There was a significant decrease in the negative control 16S bacterial load using the ddPCR technique compared with qPCR (P <0.0032). The ddPCR had a value of 0.55 ± 0.28 16S/uL and the qPCR having a value of 1.00 ± 0.70 16S copies [Figure 2A]. There was a significant positive relationship between the qPCR and ddPCR 16S counts with an R2 value of 0.27 [Figure 2B]; the line of best fit was y  =  0.33× + 1.44. Further, the ddPCR coefficients of variation (CV) were significantly lower than those obtained by the qPCR assay (P-value <0.0001) [Figure 3A]. The average CV for the ddPCR was 0.18 ± 0.14 while for the same samples the CV for the qPCR was 0.62 ± 0.29. Using a Bland-Altman plot to further analyze the CV data and using the ddPCR as the reference against the qPCR the bias was found to be −0.44 ± 0.29. This means that on average for any given sample the qPCR CV will be 0.44 ± 0.29 higher than the ddPCR CV [Figure 3B].


A comparison between droplet digital and quantitative PCR in the analysis of bacterial 16S load in lung tissue samples from control and COPD GOLD 2.

Sze MA, Abbasi M, Hogg JC, Sin DD - PLoS ONE (2014)

Comparison of the Coefficients of Variation (CV) between QPCR and ddPCR.A) Direct comparison of the ddPCR CV against the QPCR CV. The ddPCR CV was significantly lower than those obtained using qPCR. B) A Bland-Altman plot of the ddPCR CV against the qPCR CV. On average there was a much larger CV for the qPCR than the ddPCR for each individual sample of 0.44 ± 0.29.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4199711&req=5

pone-0110351-g003: Comparison of the Coefficients of Variation (CV) between QPCR and ddPCR.A) Direct comparison of the ddPCR CV against the QPCR CV. The ddPCR CV was significantly lower than those obtained using qPCR. B) A Bland-Altman plot of the ddPCR CV against the qPCR CV. On average there was a much larger CV for the qPCR than the ddPCR for each individual sample of 0.44 ± 0.29.
Mentions: The ddPCR negative controls had a much smaller standard deviation versus the qPCR negative controls (0.28 versus 0.70). Both the qPCR and ddPCR detected a similar bacterial load for the control and moderate COPD groups. For the moderate COPD group, qPCR values were 2.32 ± 0.67 16S copies (mean ± SD) and ddPCR values were 2.80 ± 1.80 16S/uL and for the control group the qPCR and ddPCR values were 2.25 ± 1.55 16S copies and 2.36 ± 1.95 16S/uL (mean ± SD) respectively. There was a significant decrease in the negative control 16S bacterial load using the ddPCR technique compared with qPCR (P <0.0032). The ddPCR had a value of 0.55 ± 0.28 16S/uL and the qPCR having a value of 1.00 ± 0.70 16S copies [Figure 2A]. There was a significant positive relationship between the qPCR and ddPCR 16S counts with an R2 value of 0.27 [Figure 2B]; the line of best fit was y  =  0.33× + 1.44. Further, the ddPCR coefficients of variation (CV) were significantly lower than those obtained by the qPCR assay (P-value <0.0001) [Figure 3A]. The average CV for the ddPCR was 0.18 ± 0.14 while for the same samples the CV for the qPCR was 0.62 ± 0.29. Using a Bland-Altman plot to further analyze the CV data and using the ddPCR as the reference against the qPCR the bias was found to be −0.44 ± 0.29. This means that on average for any given sample the qPCR CV will be 0.44 ± 0.29 higher than the ddPCR CV [Figure 3B].

Bottom Line: We hypothesize that ddPCR is better at quantifying the total bacterial load in lung tissue versus qPCR.Total 16S counts were compared between the two methods.There was no difference in the average total 16S counts (P>0.05) between the two methods.

View Article: PubMed Central - PubMed

Affiliation: Centre for Heart Lung Innovation, St. Paul's Hospital, Departments of Medicine and Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC, Canada.

ABSTRACT

Background: Low biomass in the bacterial lung tissue microbiome utilizes quantitative PCR (qPCR) 16S bacterial assays at their limit of detection. New technology like droplet digital PCR (ddPCR) could allow for higher sensitivity and accuracy of quantification. These attributes are needed if specific bacteria within the bacterial lung tissue microbiome are to be evaluated as potential contributors to diseases such as chronic obstructive pulmonary disease (COPD). We hypothesize that ddPCR is better at quantifying the total bacterial load in lung tissue versus qPCR.

Methods: Control (n = 16) and COPD GOLD 2 (n = 16) tissue samples were obtained from patients who underwent lung resection surgery, were cut on a cryotome, and sections were assigned for use in quantitative histology or for DNA extraction. qPCR and ddPCR were performed on these samples using primers spanning the V2 region on the 16S rRNA gene along with negative controls. Total 16S counts were compared between the two methods. Both methods were assessed for correlations with quantitative histology measurements of the tissue.

Results: There was no difference in the average total 16S counts (P>0.05) between the two methods. However, the negative controls contained significantly lower counts in the ddPCR (0.55 ± 0.28 16S/uL) than in the qPCR assay (1.00 ± 0.70 16S copies) (P <0.05). The coefficient of variation was significantly lower for the ddPCR assay (0.18 ± 0.14) versus the qPCR assay (0.62 ± 0.29) (P<0.05).

Conclusion: Overall the ddPCR 16S assay performed better by reducing the background noise in 16S of the negative controls compared with 16S qPCR assay.

Show MeSH
Related in: MedlinePlus