Skip to main content

Table 2 Error rates obtained according to the level of missing information in the APOE and simulated datasets

From: Computation of haplotypes on SNPs subsets: advantage of the "global method"

APOE

MD

Method

IER

SimER

IF

Res rate

0%

Global

1.94%

0.45%

0.986

100%

 

Local

4.88%

1.22%

0.97

100%

2%

Global

2.24%

0.48%

0.986

100%

 

Local

5.12%

1.19%

0.972

100%

5%

Global

3.20%

0.54%

0.987

100%

 

Local

5.64%

1.83%

0.978

100%

10%

Global

4.41%

0.68%

0.979

100%

 

Local

6.89%

1.91%

0.972

100%

15%

Global

6.98%

1.09%

0.974

100%

 

Local

10.33%

1.97%

0.964

99.75%

20%

Global

12.35%

2.09%

0.954

100%

 

Local

15.21%

2.54%

0.943

99.21%

Simulated

MD

Method

IER

SimER

IF

Res rate

0%

Global

0.14%

0.02%

0.996

100%

 

Local

0.81%

0.10%

0.989

100%

2%

Global

0.19%

0.02%

0.989

100%

 

Local

1.26%

0.13%

0.982

100%

5%

Global

0.25%

0.03%

0.982

100%

 

Local

1.46%

0.18%

0.975

100%

10%

Global

0.46%

0.05%

0.968

100%

 

Local

2.65%

0.32%

0.961

100%

15%

Global

0.83%

0.09%

0.954

100%

 

Local

4.80%

0.59%

0.947

100%

20%

Global

1.51%

0.16%

0.941

100%

 

Local

8.70%

1.06%

0.934

100%

  1. Summary of the average error rates found when working with subhaplotypes based on 4 SNPs out of 9 in the APOE genomic dataset and 10 SNPs out of 30 in the 10 simulated datasets.
  2. IER: individual error rate, Res Rate: resolution rate, SimER: similarity error rate, MD: Missing data, IF: frequency error rate (see Material and Methods).