Limits...
A flexible Bayesian method for detecting allelic imbalance in RNA-seq data.

León-Novelo LG, McIntyre LM, Fear JM, Graze RM - BMC Genomics (2014)

Bottom Line: The proposed model always has a lower type I error rate compared to the binomial test.Consequently, as variant identification improves, the need for DNA controls will be reduced.Filtering does not significantly improve performance and is not recommended, as information is sacrificed without a measurable gain.

View Article: PubMed Central - PubMed

Affiliation: Department of Biological Sciences, Auburn University, 101 Rouse Life Science Building, 36849 Auburn, AL, USA. rmgraze@auburn.edu.

ABSTRACT

Background: One method of identifying cis regulatory differences is to analyze allele-specific expression (ASE) and identify cases of allelic imbalance (AI). RNA-seq is the most common way to measure ASE and a binomial test is often applied to determine statistical significance of AI. This implicitly assumes that there is no bias in estimation of AI. However, bias has been found to result from multiple factors including: genome ambiguity, reference quality, the mapping algorithm, and biases in the sequencing process. Two alternative approaches have been developed to handle bias: adjusting for bias using a statistical model and filtering regions of the genome suspected of harboring bias. Existing statistical models which account for bias rely on information from DNA controls, which can be cost prohibitive for large intraspecific studies. In contrast, data filtering is inexpensive and straightforward, but necessarily involves sacrificing a portion of the data.

Results: Here we propose a flexible Bayesian model for analysis of AI, which accounts for bias and can be implemented without DNA controls. In lieu of DNA controls, this Poisson-Gamma (PG) model uses an estimate of bias from simulations. The proposed model always has a lower type I error rate compared to the binomial test. Consistent with prior studies, bias dramatically affects the type I error rate. All of the tested models are sensitive to misspecification of bias. The closer the estimate of bias is to the true underlying bias, the lower the type I error rate. Correct estimates of bias result in a level alpha test.

Conclusions: To improve the assessment of AI, some forms of systematic error (e.g., map bias) can be identified using simulation. The resulting estimates of bias can be used to correct for bias in the PG model, without data filtering. Other sources of bias (e.g., unidentified variant calls) can be easily captured by DNA controls, but are missed by common filtering approaches. Consequently, as variant identification improves, the need for DNA controls will be reduced. Filtering does not significantly improve performance and is not recommended, as information is sacrificed without a measurable gain. The PG model developed here performs well when bias is known, or slightly misspecified. The model is flexible and can accommodate differences in experimental design and bias estimation.

Show MeSH

Related in: MedlinePlus

Comparison of the estimatedθ. The proportion of reads coming from the paternal allele when q = DNAcontrols, as compared to q = simulation(A) and q = 1/2(B). The n = 617 exons with simulated q ≠ 0,1 and /q - 1/2/ > 0.2 are shown.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4230747&req=5

Fig4: Comparison of the estimatedθ. The proportion of reads coming from the paternal allele when q = DNAcontrols, as compared to q = simulation(A) and q = 1/2(B). The n = 617 exons with simulated q ≠ 0,1 and /q - 1/2/ > 0.2 are shown.

Mentions: Read simulation and alignment generally produce smaller estimates of bias than the DNA controls. Often the simulation results in estimates of a half even when the DNA indicates bias. However, using q from the simulation study (q=simulation) does identify a portion of the bias and only rarely does this measure estimate a larger amount of bias then the DNA. This indicates that the PG model with q = simulation should perform better than the PG model with q = 1/2 (Figure 4). Using the DNA controls as “truth”, the proportion of false positives is notably smaller and the number of false positives and false negatives are more balanced in the PG model with q = simulation, relative to the PG model with q = 1/2 (Table 2). The specificity using q = simulation is larger (0.74) than when using q = 1/2 (0.41), but the sensitivity using q = simulation is smaller (0.69) than when q = 1/2 (0.81). Among biased exonic regions there is an exorbitant false positive rate. The false positive rate (FP) is equal to 0.59 (q = 1/2), similar to what we observed in analysis of simulated RNA-seq data sets. This is substantially better than binomial (FP = 0.69), but still indicates considerable unaccounted for bias. In comparison, the false positive rate is 0.26 when q = simulation indicating that using simulated reads to estimate map bias and incorporating this measure into the statistical model dramatically reduces the false positive rate.Figure 4


A flexible Bayesian method for detecting allelic imbalance in RNA-seq data.

León-Novelo LG, McIntyre LM, Fear JM, Graze RM - BMC Genomics (2014)

Comparison of the estimatedθ. The proportion of reads coming from the paternal allele when q = DNAcontrols, as compared to q = simulation(A) and q = 1/2(B). The n = 617 exons with simulated q ≠ 0,1 and /q - 1/2/ > 0.2 are shown.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4230747&req=5

Fig4: Comparison of the estimatedθ. The proportion of reads coming from the paternal allele when q = DNAcontrols, as compared to q = simulation(A) and q = 1/2(B). The n = 617 exons with simulated q ≠ 0,1 and /q - 1/2/ > 0.2 are shown.
Mentions: Read simulation and alignment generally produce smaller estimates of bias than the DNA controls. Often the simulation results in estimates of a half even when the DNA indicates bias. However, using q from the simulation study (q=simulation) does identify a portion of the bias and only rarely does this measure estimate a larger amount of bias then the DNA. This indicates that the PG model with q = simulation should perform better than the PG model with q = 1/2 (Figure 4). Using the DNA controls as “truth”, the proportion of false positives is notably smaller and the number of false positives and false negatives are more balanced in the PG model with q = simulation, relative to the PG model with q = 1/2 (Table 2). The specificity using q = simulation is larger (0.74) than when using q = 1/2 (0.41), but the sensitivity using q = simulation is smaller (0.69) than when q = 1/2 (0.81). Among biased exonic regions there is an exorbitant false positive rate. The false positive rate (FP) is equal to 0.59 (q = 1/2), similar to what we observed in analysis of simulated RNA-seq data sets. This is substantially better than binomial (FP = 0.69), but still indicates considerable unaccounted for bias. In comparison, the false positive rate is 0.26 when q = simulation indicating that using simulated reads to estimate map bias and incorporating this measure into the statistical model dramatically reduces the false positive rate.Figure 4

Bottom Line: The proposed model always has a lower type I error rate compared to the binomial test.Consequently, as variant identification improves, the need for DNA controls will be reduced.Filtering does not significantly improve performance and is not recommended, as information is sacrificed without a measurable gain.

View Article: PubMed Central - PubMed

Affiliation: Department of Biological Sciences, Auburn University, 101 Rouse Life Science Building, 36849 Auburn, AL, USA. rmgraze@auburn.edu.

ABSTRACT

Background: One method of identifying cis regulatory differences is to analyze allele-specific expression (ASE) and identify cases of allelic imbalance (AI). RNA-seq is the most common way to measure ASE and a binomial test is often applied to determine statistical significance of AI. This implicitly assumes that there is no bias in estimation of AI. However, bias has been found to result from multiple factors including: genome ambiguity, reference quality, the mapping algorithm, and biases in the sequencing process. Two alternative approaches have been developed to handle bias: adjusting for bias using a statistical model and filtering regions of the genome suspected of harboring bias. Existing statistical models which account for bias rely on information from DNA controls, which can be cost prohibitive for large intraspecific studies. In contrast, data filtering is inexpensive and straightforward, but necessarily involves sacrificing a portion of the data.

Results: Here we propose a flexible Bayesian model for analysis of AI, which accounts for bias and can be implemented without DNA controls. In lieu of DNA controls, this Poisson-Gamma (PG) model uses an estimate of bias from simulations. The proposed model always has a lower type I error rate compared to the binomial test. Consistent with prior studies, bias dramatically affects the type I error rate. All of the tested models are sensitive to misspecification of bias. The closer the estimate of bias is to the true underlying bias, the lower the type I error rate. Correct estimates of bias result in a level alpha test.

Conclusions: To improve the assessment of AI, some forms of systematic error (e.g., map bias) can be identified using simulation. The resulting estimates of bias can be used to correct for bias in the PG model, without data filtering. Other sources of bias (e.g., unidentified variant calls) can be easily captured by DNA controls, but are missed by common filtering approaches. Consequently, as variant identification improves, the need for DNA controls will be reduced. Filtering does not significantly improve performance and is not recommended, as information is sacrificed without a measurable gain. The PG model developed here performs well when bias is known, or slightly misspecified. The model is flexible and can accommodate differences in experimental design and bias estimation.

Show MeSH
Related in: MedlinePlus