Commit d4f1e790 authored by Jakubek's avatar Jakubek
Browse files

updated FAQ

parent 75445619
RECUR Usage FAQ
## RECUR Usage FAQ
What genome builds are supported?
The default is set to hg19, but there is an option to provide your own reference file with chromosome lengths and centromere positions for other genome builds.
What types of data can I analyze?
You can use data from both snp and next-gen sequencing platforms including targeted sequencing. The number and distribution of heterozygous markers will dictate the power for the detection of bidirectional allelic imbalance.
What types of data can I analyze?
Can I use this algorithm on other species?
You can use data from both snp and next-gen sequencing platforms including targeted sequencing. The number and distribution of heterozygous markers typed by the specific method will dictate the power for detection of bidirectional allelic imbalance. Small targeted panels may not provide enough heterozygous for analysis. We have applied this method to whole exome sequencing and SNP array data (Illumina and Affymetrix).
How does percent tumor cells affect RECUR?
The fraction of cells that carry the chromosomal alteration with be positively correlated with the power to detect mirrored AI events.
Can you use this algorithm on subtle AI events?
Yes, we habe applied the algorithm to AI events that could not be classified as a gain, loss, or CNLOH event. Other factors that can affect the analysis include event length and the BAF deviations in all the samples from the same individual.
Can I use this algorithm on data from other species (no human)?
Yes, you would have to provide a genome reference file with chromosome lengths and centromere boundaries.
How does the user denote missing genotype and BAF/LRR values?
......@@ -18,6 +25,9 @@ Set to 0.5 and 0 if no predefined AI segments are present; else they are calcula
My files have homozygous markers, can I still use RECUR?
Yes, the algorithm ignores homozygous markers.
How can I change plotting parameters?
You can edit the mirror_ai_plot function in the R code.
How does the window testing option work? How many markers should I use?
This option is best for refining AI profiles in a specific genomic segment, such as a chromosome arm.
......@@ -25,5 +35,3 @@ The window testing option breaks AI segments into windows of N markers (where N
We suggest using 20 or more markers for the window testing as a smaller number will decrease statistical power for detection of bidirectional AI. All statistically significant windows are reported, therefore, it may be best to test 500-100 marker windows first. Small window sizes and large AI segments for testing will slow down the algorithm. The window option may report a very large number of bidirectional events in cases where there are several samples and/or large event(s).
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment