Genotype is not available at position # ########

[andhena]

Dear Paul
I have another question about the One_step_clustering() function.
I ran Control-FREEC on one sample and tried QuantumClone.
Here is what my input SNV file looks like:

head(SNV_file)

  SampleName Chr   Start Depth Alt
1 sample1   1  909419    16   3
2 sample1   1  986894    20   2
3 sample1   1 1262354    95   3
4 sample1   1 1563476    73   2
5 sample1   1 1585590   117   2
6 sample1   1 1654193   247   3

And what the output of Control-FREEC looks like for this sample:

head(CNV_file)

Chromosome  Start    Ratio MedianRatio CopyNumber BAF estimatedBAF Genotype UncertaintyOfGT              Gene
1          1  65535 0.726426    0.508169          1  -1            1        A              -1   1: 65534- 65600
2          1  65857  1.12502    0.508169          1  -1            1        A              -1   1: 65856- 65948
3          1  69091 0.851791    0.508169          1   1            1        A              -1   1: 69090- 70008
4          1 721407 0.524401    0.508169          1  -1            1        A              -1 1: 721406- 721494
5          1 721556 0.876531    0.508169          1  -1            1        A              -1 1: 721555- 721781
6          1 752751  1.17044    0.508169          1   1            1        A              -1 1: 752750- 753582

When I tried the following commands:

SNV_liste = list(SNV_file)
FREEC_liste = list(CNV_file)
contamination = 30
One_step_clustering(SNV_list = SNV_liste, FREEC_list = FREEC_liste, contamination, nclone_range = 2:5, clone_priors = NULL, prior_weight = NULL, Initializations = 1, preclustering = "FLASH", simulated = FALSE, epsilon = NULL, save_plot = TRUE, ncores = 1, restrict.to.AB = FALSE, output_directory = NULL, model.selection = "BIC", optim = "default", keep.all.models = FALSE, force.single.copy = FALSE)

I got:

Checking that SNV_list and FREEC_list have the same number of samples...
Passed
epsilon set to: 0.00566125684264668
Checking all possibilities for sample1
Genotype is not available at position 1 909419
Genotype is not available at position 1 986894
Genotype is not available at position 1 1262354
[...]
Genotype is not available at position 21 47832900
Error in lis[[1]] : subscript out of bounds
In addition: Warning message:
In Patient_schrodinger_cellularities(SNV_list = SNV_list, FREEC_list = FREEC_list,  :
  4931 mutations excluded due to missing genotype or normalization issues

Could you please explain me what I am not doing right?
I guess I could add a genotype column in my snv file but I was hoping I could directly use the output file sample1_ratio.txt from Control-FREEC.
Thank you in advance

Kind Regards
-Charlotte

[DeveauP]

Hi Charlotte,

First thing, contamination should be between 0 and 1, so using a contamination of 0.3 would reduce the number of variants rejected due to normalization issues.
Second, could you please check that the genotype is not -1 around your position (the line with the Start strictly lower to the variant you are looking at)?

Best wishes,
Paul

[andhena]

Hi Paul,

Yes indeed, I made a mistake contamination value when I wrote the issue (I noticed as soon as I posted the comment). I actually tested contamination = 0 when I tried to run QuantumClone.

When I check the genotype of the 6 variants detected, I got for instance A, and for UncertaintyOfGT I got -1 :

SNV file

  SampleName Chr   Start Depth Alt
1 sample1   1  909419    16   3
2 sample1   1  986894    20   2
3 sample1   1 1262354    95   3
4 sample1   1 1563476    73   2
5 sample1   1 1585590   117   2
6 sample1   1 1654193   247   3

CNV file

Chromosome	Start	Ratio	MedianRatio	CopyNumber	BAF	estimatedBAF	Genotype	UncertaintyOfGT	Gene
1	909213	0.505172	0.508169	1	1	1	A	-1	1: 909212- 909431
1	986833	0.593654	0.508169	1	-1	1	A	-1	1: 986832- 987025
1	1262621	0.599795	0.508169	1	-1	1	A	-1	1: 1262620- 1263143
1	1563399	0.566113	0.508169	1	1	1	A	-1	1: 1563398- 1563559
1	1585560	0.660535	0.508169	1	1	1	A	-1	1: 1585559- 1585739
1	1654147	0.996888	0.508169	1	-1	1	A	-1	1: 1654146- 1654257

Do you have an idea of what I should do to make it work?

Kind Regards,
-Charlotte

[DeveauP]

Hi Charlotte,

To prevent any delay in your work, I think the quickest would be to insert the genotype directly in your SNV table.
I'll try to fix the FREEC concatenation as soon as I can using the data you provided, but it may not be before next week.

Best wishes,
Paul

[andhena]

Hi Paul,

Thank you for your answer and your help.
I'll add the genotype in the SNV table in the meantime.

Kind Regards,
-Charlotte

[DeveauP]

Hi Charlotte,

The QuantumClone version 1.0.0.7 available on Github should fix your issue.

Best wishes,
Paul

[andhena]

Hi Paul,
Thanks for your doing it so quickly.
Unfortunately I have a new error now :

Checking that SNV_list and FREEC_list have the same number of samples...
Passed
epsilon set to: 0.00566125684264668
Checking all possibilities for <my sample>
Error in `[.data.frame`(z, as.character(z[, "Chromosome"]) == as.character(chr[i])) : 
  undefined columns selected

I tried different things to fix it, including renaming columns but errors are persisting, do you know where this problem comes from?

-Charlotte

[DeveauP]

Hi Charlotte,

Not having any FREEC file example, I worked on your extract, and used the first line as header. Could you check that there is a "Chromosome" column name?

Best wishes,
Paul

[andhena]

Hi Paul,

I'm using the same file, and the column name is "Chromosome" indeed:

> colnames(CNV_file)
 [1] "Chromosome"      "Start"           "Ratio"           "MedianRatio"     "CopyNumber"     
 [6] "BAF"             "estimatedBAF"    "Genotype"        "UncertaintyOfGT" "Gene"   

> head(CNV_file$Chromosome)
[1] 1 1 1 1 1 1
Levels: 1 10 11 12 13 14 15 16 17 18 19 2 20 21 22 3 4 5 6 7 8 9 X

> mode(CNV_file$Chromosome)
[1] "numeric"

Thank you again for your help,

Kind regards
-Charlotte

[DeveauP]

Hi Charlotte,

I tried a quick fix, tell me if it works (current version on Github).

Paul

[andhena]

Hi Paul,

Still not working, I've got something slightly different :

Error in `[.data.frame`(z, as.character(z[, ChrCol]) == as.character(chr[i])) : 
  undefined columns selected

-Charlotte

[DeveauP]

Hi Charlotte,

It seems that object received by the function is not a data.frame anymore. Are you still passing FREEC_list = FREEC_liste as argument or FREEC_list = CNV_file?

Best wishes,
Paul

[andhena]

Hi Paul,

I'm still passing

SNV_liste = list(SNV_file)
FREEC_liste = list(CNV_file)
contamination = 0
One_step_clustering(SNV_list = SNV_liste, FREEC_list = FREEC_liste, contamination, nclone_range = 2:5, clone_priors = NULL, prior_weight = NULL, Initializations = 1, preclustering = "FLASH", simulated = FALSE, epsilon = NULL, save_plot = TRUE, ncores = 1, restrict.to.AB = FALSE, output_directory = NULL, model.selection = "BIC", optim = "default", keep.all.models = FALSE, force.single.copy = FALSE)

Then I tried FREEC_list = CNV_file just in case but it's not working.

Kind regards,

-Charlotte

[DeveauP]

Hi Charlotte,

I have been trying to reproduce your error using your file, but to no avail. I cannot figure out what triggers your error as it works on my computer.
Unless you can provide me with a reproducible example, I'll mark this as closed.

Paul

[andhena]

Hi Paul,
I sent you an email last week with my input files and the complete script I used at quantumclone.package at gmail.com
Thank you for your help and your time,
Best
-Charlotte