Applied Biosystems SOLiD™ 4 System SETS Software User Guide ...

Sampling
Allele ratio
Applied Biosystems SOLiD 4 System SETS Software User Guide
Appendix B Advanced Topic: Data Analysis Overview
Data analysis considerations B
If coverage at a genome position is low, the second allele might often
not be sampled, even if it is present. Heterozygotes are expected to be
underrepresented at low coverage. (They can be called as a
homozygote for one of the two alleles.) It is important to have a low
false positive rate for SNP detection, because SNPs are expected to
exist at only 1 in 1,000 positions for humans and some other species.
The false positive rate in an individual species should be at least an
order of magnitude lower than this, to avoid a high false discovery
rate. An error model that incorporates quality values (QV) can
increase the overall accuracy of SNP detection. The SNP detection
algorithms of the SOLiD 4 System use explicit error models
generated from each run and biologically meaningful prior
probabilities to evaluate evidence of a SNP. 2-base encoding provides
a built-in way to distinguish errors from true alleles to manage the
false positive rate.
If the sample preparation method or some other cause produces
results in allele ratios that are considerably different from the
expected 50:50 ratio, it is more difficult to detect heterozygosity.
Detecting heterozygosity then requires more sequence coverage.
Find polymorphisms in color-space
The SOLiD 4 System can detect complicated genomic variations
such as adjacent SNPs, insertions, deletions, and structural
rearrangements.
For other analysis tools, visit the AB SOLiD Software Community
website:
http://www3.appliedbiosystems.com/AB_Home/applicationstechnol
ogies/SOLiDSystemSequencing/SoftwareCommunityDataAnalysis
ResourcesforScientistsDevelopers/index.htm
An example of various polymorphisms in color-space is shown in the
figure below.
155