How To Interpret The Output Generated By The Calc-Bmr Music By Washu

Question

[Ashwini Patil]

hi,

I have 7 genes from 1 sample I am trying to analyze. In the result, the total no. of mutations for every gene is shown as the total no. of genes I have provide in the maf file.

#Gene    Indels    SNVs    Tot Muts
MARCH12 0    7    7 
JUNE12    0    7    7
....

Here, for every gene it shows 7 mutations. Is this what it is supposed to give? I think I am going wrong somewhere. I have tried giving different no. of genes. If i give 11 genes in maf file it shows 11 mutations for each and every gene. I understand that the mutation it is showing are the mutations that I have given in the maf file but giving that same no. for every gene is something I am doubtful off. Please help.

Ashwini

[Cyriac Kandoth]

Can you revise your question a bit? For the example you provided, did you mean to say - "if there are 7 samples in maf file"...? Also, note that Biostar uses the Markdown language i.e. if you use two underscores around some text, the text will be italics, and the underscores will disappear. To keep this from happening, type a backslash before each underscore.

Answer 1

[Cyriac Kandoth]

There is likely something incorrect in either your MAF file or the ROI file. I see that your gene names are MARCH12 and JUNE12 which indicates that you've used Excel at some point. Excel has a habit of converting gene names into calendar days. There's a whole paper about this, so I recommend using gnumeric instead. Please also ensure that the MAF file uses UNIX newline characters, like with the ROI file. If you need sample ROI files, you can find them here. And click here for sample MAF files.

Each line in the MAF file should list a single mutation from a single sample. Column 1 of each line specifies the gene that the mutation alters, and column 16 specifies the sample ID. The SMG test is most effective when you have mutations from many samples, preferably hundreds of cases from the disease being studied. Columns 9 and 10 specify the variant type, where Silent mutations are skipped by default in calc-bmr. Full column MAF specs are at this link, but calc-bmr only cares that you have valid values in these columns:

1. Hugo_Symbol
5. Chromosome
6. Start_Position
7. End_Position
9. Variant_Classification
10. Variant_Type
11. Reference_Allele
12. Tumor_Seq_Allele1
13. Tumor_Seq_Allele2
16. Tumor_Sample_Barcode

[Ashwini Patil]

hi, I am sorry to come back to this so late. I checked there is no problem with other files. The thing is that I have only 1 mutation per gene mentioned in the maf file. If we see the above example, there are 7 mutations listed in maf file. In the results from bmr it lists 7 mutations for each gene instead of 1 and the total no. of mutation hence comes out to be 7*7=49. So the bmr calculated is 7 times of what it should be. This may not be affecting the final output from smg but I am not able to understand why it shows no. of mutation per gene to be 7 instead of 1.

[Cyriac Kandoth]

Please point us to some sample inputs to reproduce your error, and we can proceed from there.

Answer 2

[Chris Miller]

I'm not sure I entirely understand your question, but there doesn't appear to be anything wrong with the output you posted. It indicates that you have 7 mutations in LATS1, all SNVs. If you had 2 indels and 5 SNVs, the total mutations would still be 7. According to the FDR values, this gene is mutated more frequently than would be expected, based on the background mutation rate in the samples.

[Ashwini Patil]

Thank you so much for the quick reply. I am extermely sorry to be so clumsy. I have revised my question.