Hello,
first of all, thank you very much for providing this tool.
I have a set of male control and tumor samples generated by WES. I have run batch command and I've seen in the log that some samples sex has been inferred incorrectly.
On one side, all control samples seem to be inferred correctly when using target coverages (all assigned as males) but most of them are assigned as females when using antitarget coverages. Then the algorithm prints "Sample X chromosomal X/Y ploidy looks like male in targets but female in antitargets; preferring antitargets", from which I understand that it is assigning the sample as female, which is incorrect.
On the other side, some of the tumor samples are also assigned incorrectly as female, but only one inference is printed for each sample, while for the control samples we get two inferences, one from targets and one from antitargets.
I've been going through the documentation but I'm still not sure how the algorithm works and how a wrong sex inference can influence the results of X chromosome. My specific questions are the following:
1- In which step does the batch command infer the sex of the samples, both controls used to build the reference and tumor samples?
2- Why when the inference from targets and antitargets doesn't match, the program prefers antitargets?
3- How does the sex inferred of the tumor samples affect the results of chrX? Does it affect the CN column in *call.cns file or it affects only the diagram (for example a CN=2, is represented as neutral in the diagram if the sample is assigned as female, and as a gain if the sample is assigned as male)?
4- How does the sex inferred of the control samples affect the results of chrX?
5- I've seen that some commands allow assigning the sex of the samples with the option -x. Does the batch command include it?
Hello, first of all, thank you very much for providing this tool.
I have a set of male control and tumor samples generated by WES. I have run batch command and I've seen in the log that some samples sex has been inferred incorrectly. On one side, all control samples seem to be inferred correctly when using target coverages (all assigned as males) but most of them are assigned as females when using antitarget coverages. Then the algorithm prints "Sample X chromosomal X/Y ploidy looks like male in targets but female in antitargets; preferring antitargets", from which I understand that it is assigning the sample as female, which is incorrect. On the other side, some of the tumor samples are also assigned incorrectly as female, but only one inference is printed for each sample, while for the control samples we get two inferences, one from targets and one from antitargets. I've been going through the documentation but I'm still not sure how the algorithm works and how a wrong sex inference can influence the results of X chromosome. My specific questions are the following:
1- In which step does the batch command infer the sex of the samples, both controls used to build the reference and tumor samples? 2- Why when the inference from targets and antitargets doesn't match, the program prefers antitargets? 3- How does the sex inferred of the tumor samples affect the results of chrX? Does it affect the CN column in *call.cns file or it affects only the diagram (for example a CN=2, is represented as neutral in the diagram if the sample is assigned as female, and as a gain if the sample is assigned as male)? 4- How does the sex inferred of the control samples affect the results of chrX? 5- I've seen that some commands allow assigning the sex of the samples with the option -x. Does the batch command include it?
Thank you very much,
Alba