Skip to main content

Table 1 Parameter Estimates for the Exponential-Normal Convolution Model

From: Error correction and statistical analyses for intra-host comparisons of feline immunodeficiency virus diversity from high-throughput sequencing data

Library pairs

\( {\widehat{\alpha}}^a \)

\( {\widehat{\mu}}^b \)

\( {\widehat{\sigma}}^c \)

Total Reads

Average Coverage

1 & 2

637

3.2×10-4

5.1×10-5

63439

1810

3 & 4

750

2.4×10-4

4.7×10-5

90839

2775

5 & 6

810

1.4×10-4

1.7×10-5

133801

4236

7 & 8

813

1.8×10-4

2.1×10-5

132589

4078

9 & 10

865

1.4×10-4

3.6×10-5

177873

5530

11 & 12

888

2.1×10-4

6.6×10-5

118396

3916

All

777

1.6×10-4

5.1×10-5

716937

3724

  1. a \( \widehat{\alpha} \): estimated rate parameter for the exponentially distributed signal (obtained anchoring the 75th percentile of an exponential distribution to the 75th percentile of all frequencies minus the estimate of μ (i.e. the overall mode); b \( \widehat{\mu} \): estimated mean for the normally distributed error (fixed at the overall mode); c \( \widehat{\sigma} \): estimated standard deviation for the normally distributed error (obtained “doubling” the spread on the left of the overall mode)