Limits...
The Power Decoder Simulator for the Evaluation of Pooled shRNA Screen Performance.

Stombaugh J, Licon A, Strezoska Ž, Stahl J, Anderson SB, Banos M, van Brabant Smith A, Birmingham A, Vermeulen A - J Biomol Screen (2015)

Bottom Line: Using the negative binomial distribution, it models both the relative abundance of multiple shRNAs within a single screening replicate and the biological noise between replicates for each individual shRNA.We demonstrate that this simulator can successfully model the data from an actual laboratory experiment.The Power Decoder simulator is written in R and Python and is available for download under the GNU General Public License v3.0.

View Article: PubMed Central - PubMed

Affiliation: Dharmacon, part of GE Healthcare, Lafayette, CO, USA.

No MeSH data available.


Differential enrichment and depletion of short hairpin RNAs (shRNAs) in engineered screens. MA plots of representative examples of normalized data from experimental shRNA pooled screens with engineered twofold enrichment and depletion of shRNAs in which transductions were performed at (A) 100 and (B) 500 independent shRNA integrations on average. The shRNAs with significantly (p* ≤ 0.05) higher and lower abundance in T1 in the next-generation sequencing count data are in red and blue, respectively. Power values listed are mean ± standard deviation over 30 normalizations.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2 - License 3
getmorefigures.php?uid=PMC4543901&req=5

fig1-1087057115576715: Differential enrichment and depletion of short hairpin RNAs (shRNAs) in engineered screens. MA plots of representative examples of normalized data from experimental shRNA pooled screens with engineered twofold enrichment and depletion of shRNAs in which transductions were performed at (A) 100 and (B) 500 independent shRNA integrations on average. The shRNAs with significantly (p* ≤ 0.05) higher and lower abundance in T1 in the next-generation sequencing count data are in red and blue, respectively. Power values listed are mean ± standard deviation over 30 normalizations.

Mentions: More true hits were identified in Screen 500_2x than in Screen 100_2x, and Screen 500_2x shows a better separation of the identified depletion and enrichment hits (Fig. 1). Of the 960 shRNAs that were manually varied in T1 by twofold, only an average of 259 (σ = 4.13) were identified as hits in the Screen 100_2x, whereas an average of 877 (σ = 3.71) were identified as hits in Screen 500_2x (Suppl. Table S3). Although the false-positive rate was less than 1.5% in both screens due to the multiple test correction, the false-negative rate decreased substantially from 72.40% (σ = 0.45) to 6.74% (σ = 0.39) as shRNA fold representation increased, giving a concomitant increase in screen power from 27.60% (σ = 0.45) to 93.26% (σ = 0.39) from Screen 100_2x to Screen 500_2x. Manually increasing the shRNA enrichment and depletion magnitude from 2- to 4-fold in Screen 100 improved the power from 27.60% (σ = 0.45) to 76.10% (σ = 0.55), and manually decreasing the shRNA enrichment and depletion from 2- to 1.5-fold in Screen 500 led to a drop in power from 93.26% (σ = 0.39) to 58.35% (σ = 0.92; Suppl. Table S3). This substantiates previous findings that strong hits can be identified at a range of shRNA fold representations whereas moderate hits may be lost at lower shRNA fold representations13,14 and provides confidence that these data sets are realistic as the basis of a simulator.


The Power Decoder Simulator for the Evaluation of Pooled shRNA Screen Performance.

Stombaugh J, Licon A, Strezoska Ž, Stahl J, Anderson SB, Banos M, van Brabant Smith A, Birmingham A, Vermeulen A - J Biomol Screen (2015)

Differential enrichment and depletion of short hairpin RNAs (shRNAs) in engineered screens. MA plots of representative examples of normalized data from experimental shRNA pooled screens with engineered twofold enrichment and depletion of shRNAs in which transductions were performed at (A) 100 and (B) 500 independent shRNA integrations on average. The shRNAs with significantly (p* ≤ 0.05) higher and lower abundance in T1 in the next-generation sequencing count data are in red and blue, respectively. Power values listed are mean ± standard deviation over 30 normalizations.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2 - License 3
Show All Figures
getmorefigures.php?uid=PMC4543901&req=5

fig1-1087057115576715: Differential enrichment and depletion of short hairpin RNAs (shRNAs) in engineered screens. MA plots of representative examples of normalized data from experimental shRNA pooled screens with engineered twofold enrichment and depletion of shRNAs in which transductions were performed at (A) 100 and (B) 500 independent shRNA integrations on average. The shRNAs with significantly (p* ≤ 0.05) higher and lower abundance in T1 in the next-generation sequencing count data are in red and blue, respectively. Power values listed are mean ± standard deviation over 30 normalizations.
Mentions: More true hits were identified in Screen 500_2x than in Screen 100_2x, and Screen 500_2x shows a better separation of the identified depletion and enrichment hits (Fig. 1). Of the 960 shRNAs that were manually varied in T1 by twofold, only an average of 259 (σ = 4.13) were identified as hits in the Screen 100_2x, whereas an average of 877 (σ = 3.71) were identified as hits in Screen 500_2x (Suppl. Table S3). Although the false-positive rate was less than 1.5% in both screens due to the multiple test correction, the false-negative rate decreased substantially from 72.40% (σ = 0.45) to 6.74% (σ = 0.39) as shRNA fold representation increased, giving a concomitant increase in screen power from 27.60% (σ = 0.45) to 93.26% (σ = 0.39) from Screen 100_2x to Screen 500_2x. Manually increasing the shRNA enrichment and depletion magnitude from 2- to 4-fold in Screen 100 improved the power from 27.60% (σ = 0.45) to 76.10% (σ = 0.55), and manually decreasing the shRNA enrichment and depletion from 2- to 1.5-fold in Screen 500 led to a drop in power from 93.26% (σ = 0.39) to 58.35% (σ = 0.92; Suppl. Table S3). This substantiates previous findings that strong hits can be identified at a range of shRNA fold representations whereas moderate hits may be lost at lower shRNA fold representations13,14 and provides confidence that these data sets are realistic as the basis of a simulator.

Bottom Line: Using the negative binomial distribution, it models both the relative abundance of multiple shRNAs within a single screening replicate and the biological noise between replicates for each individual shRNA.We demonstrate that this simulator can successfully model the data from an actual laboratory experiment.The Power Decoder simulator is written in R and Python and is available for download under the GNU General Public License v3.0.

View Article: PubMed Central - PubMed

Affiliation: Dharmacon, part of GE Healthcare, Lafayette, CO, USA.

No MeSH data available.