Handling ‘*’ Alleles in MSMC2 Input Files

[GNWEN]

Hello @stschiff ,

First of all, thank you for developing and maintaining such a great tool!

I have a question regarding how MSMC2 handles the ‘*’ allele in input files. My VCF file is original generated from the GATK pipeline and contains some ‘*’ alleles, as shown in the example below:

Chr01    126405    2855    C*  
Chr01    126850    233    C*  
Chr01    128017    449    TC  
Chr01    128686    540    GA  
Chr01    128723    37    A*  
Chr01    129065    35    AT  

In the context of GATK's HaplotypeCaller, the ‘*’ allele represents a spanning deletion, indicating that a deletion in one sample overlaps a variant site in another sample. This notation is used to maintain consistency in variant calls across multiple samples.

I am wondering how MSMC2 processes such sites. In my case, when running MSMC2 on individual samples, those with ‘*’ alleles in their input seem to run normally and produce results. However, for samples where the input starts with a ‘*’ allele (as the example I gave above), the program fails with the following error message:

error in parsing command line: object.Exception@model/data.d(57): could not parse line: Chr01      126405  2855    C*
----------------
??:? pure @safe void std.exception.bailOut!(Exception).bailOut(immutable(char)[], ulong, const(char[])) [0x52ec51]
??:? @safe std.regex.__T10RegexMatchTAxaS453std5regex8internal8thompson15ThompsonMatcherZ.RegexMatch std.exception.enforce!(Exception, std.regex.__T10RegexMatchTAxaS453std5regex8internal8thompson15ThompsonMatcherZ.RegexMatch).enforce(std.regex.__T10RegexMatchTAxaS453std5regex8internal8thompson15ThompsonMatcherZ.RegexMatch, lazy const(char)[], immutable(char)[], ulong) [0x548729]
??:? void model.data.checkDataLine(const(char[])) [0x51f3d2]
??:? ulong model.data.getNrHaplotypesFromFile(immutable(char)[]) [0x51f4f0]
??:? void msmc2.parseCommandLine(immutable(char)[][]) [0x572d41]
??:? _Dmain [0x5728d3]
??:? _D2rt6dmain211_d_run_mainUiPPaPUAAaZiZ6runAllMFZ9__lambda1MFZv [0x5cbc56]
??:? void rt.dmain2._d_run_main(int, char**, extern (C) int function(char[][])*).tryExec(scope void delegate()) [0x5cbbac]
??:? void rt.dmain2._d_run_main(int, char**, extern (C) int function(char[][])*).runAll() [0x5cbc12]
??:? void rt.dmain2._d_run_main(int, char**, extern (C) int function(char[][])*).tryExec(scope void delegate()) [0x5cbbac]
??:? _d_run_main [0x5cbb09]
??:? main [0x57bbc5]
??:? __libc_start_main [0xdb524554]

Could you please advise on the correct way to handle these ‘*’ alleles when preparing input files for MSMC2? Any guidance would be greatly appreciated!

Best regards,
Guannan

[stschiff]

Hi @weirdfanny, thanks for reaching out. MSMC cannot handle deletions. You need to filter these out and only consider true SNPs. Does that help?