ABSOLUTE MAF error

Question

[Alejandro Jimenez Sanchez]

I am running ABSOLUTE 1.0.6 in R 3.4.0

When I do:

RunAbsolute(seg.dat.fn, sigma.p, max.sigma.h, min.ploidy, max.ploidy, primary.disease, platform, sample.name, results.dir,  max.as.seg.count, max.non.clonal, max.neg.genome, copy_num_type, maf.fn, min.mut.af, output.fn.base=NULL, verbose=TRUE)

I get:

[1] "Capping 1 segs at tCR = 5.0"
[1] "Expected copy-ratio = 0.89228"
[1] "Removing 0 of 42303 mutations with NA coverage"
[1] "Removing NA of 42303 mutations due to allelic fraction < 0"
[1] "Removing 0 mutations in IG regions" 
Error in if (sum(seg.ix) != 1) { : missing value where TRUE/FALSE needed

If I do:

maf.fn <- NULL

Everything runs smoothly, so the error should be in the MAF file.

The headers of the MAF are:

   'Hugo_Symbol', 'Entrez_Gene_Id', 'Center', 'NCBI_Build',
   'Chromosome', 'Start_position', 'End_position', 'Strand',
   'Variant_Classification', 'Variant_Type', 'Reference_Allele',
   'Tumor_Seq_Allele1', 'Tumor_Seq_Allele2', 'dbSNP_RS',
   'dbSNP_Val_Status', 'Tumor_Sample_Barcode',
   'Matched_Norm_Sample_Barcode', 'Match_Norm_Seq_Allele1',
   'Match_Norm_Seq_Allele2', 'Tumor_Validation_Allele1',
   'Tumor_Validation_Allele2', 'Match_Norm_Validation_Allele1',
   'Match_Norm_Validation_Allele2', 'Verification_Status',
   'Validation_Status', 'Mutation_Status', 'Sequencing_Phase',
   'Sequence_Source', 'Validation_Method', 'Score', 'BAM_File',
   'Sequencer', 'Tumor_Sample_UUID', 'Matched_Norm_Sample_UUID',
   'HGVSc', 'HGVSp', 'HGVSp_Short', 'Transcript_ID', 'Exon_Number',
   't_depth', 't_ref_count', 't_alt_count', 'n_depth', 'n_ref_count',
   'n_alt_count', 'all_effects', 'Allele', 'Gene', 'Feature',
   'Feature_type', 'Consequence', 'cDNA_position', 'CDS_position',
   'Protein_position', 'Amino_acids', 'Codons', 'Existing_variation',
   'ALLELE_NUM', 'DISTANCE', 'STRAND_VEP', 'SYMBOL', 'SYMBOL_SOURCE',
   'HGNC_ID', 'BIOTYPE', 'CANONICAL', 'CCDS', 'ENSP', 'SWISSPROT',
   'TREMBL', 'UNIPARC', 'RefSeq', 'SIFT', 'PolyPhen', 'EXON', 'INTRON',
   'DOMAINS', 'GMAF', 'AFR_MAF', 'AMR_MAF', 'ASN_MAF', 'EAS_MAF',
   'EUR_MAF', 'SAS_MAF', 'AA_MAF', 'EA_MAF', 'CLIN_SIG', 'SOMATIC',
   'PUBMED', 'MOTIF_NAME', 'MOTIF_POS', 'HIGH_INF_POS',
   'MOTIF_SCORE_CHANGE', 'IMPACT', 'PICK', 'VARIANT_CLASS', 'TSL',
   'HGVS_OFFSET', 'PHENO', 'MINIMISED', 'ExAC_AF', 'ExAC_AF_AFR',
   'ExAC_AF_AMR', 'ExAC_AF_EAS', 'ExAC_AF_FIN', 'ExAC_AF_NFE',
   'ExAC_AF_OTH', 'ExAC_AF_SAS', 'GENE_PHENO', 'FILTER',
   'flanking_bps', 'variant_id', 'variant_qual', 'ExAC_AF_Adj',
   'ExAC_AC_AN_Adj', 'ExAC_AC_AN', 'ExAC_AC_AN_AFR', 'ExAC_AC_AN_AMR',
   'ExAC_AC_AN_EAS', 'ExAC_AC_AN_FIN', 'ExAC_AC_AN_NFE',
   'ExAC_AC_AN_OTH', 'ExAC_AC_AN_SAS', 'ExAC_FILTER'

I want to run ABSOLUTE with the MAF to include the SNV information. Does anyone know how to solve the error generated when the MAF file is included?

Thanks, Alejandro

[poisonAlien]

Could be wrong, but I think p in Start_position and End_position should start with uppercase. Start_Position End_Position

[Alejandro Jimenez Sanchez]

Thanks for your comment. Indeed the original MAF file had it with uppercase, but when I run it I get the following error:

Error in `[.data.frame`(maf, , "Start_position") : 
undefined columns selected

Thus, I changed Start_Position to Start_position, the same for End_position. Then I get the error:

Error in if (sum(seg.ix) != 1) { : missing value where TRUE/FALSE needed

Which is the question posted. If you have any other idea why this error may be, please let me know. :)

[poisonAlien]

Absolute requires column containing vaf, t_vaf. Maybe try adding it ?

maf$t_vaf = as.numeric(as.character(maf$t_alt_count)) / as.numeric(as.character(maf$t_depth))

edit: also make sure chromosome names in MAF matches to that in segmentatioin file. Better remove chr prefix from both files.

[Alejandro Jimenez Sanchez]

I added t_vaf and make sure that the chromosome names were the same between the MAF and the seg file. I should say that the data I am trying to analyse is mouse data, and when I run human data I don't have the problem. I know Absolute was developed for human data, but I was giving it a try. I want to get absolute copy numbers, purity, and clonality of this mouse data, so I was thinking in another tool such as FACETS, which can use mouse data.

Best, Alejandro

[poisonAlien]

oh ! Okay, I think Absolute uses pre-computed purity and ploidy models from known cancer cohorts. I guess this is an issue if you're using non-human genome.

[emilie.lalonde]

Hello,

Were you able to resolve this issue?

I am having the same error using human cancer samples on Absolute v1.0.6. I also adjusted my MAF column headers, added the t_vaf column, and removed the 'chr' prefix for the Chromosome column.

Thanks!

[Alejandro Jimenez Sanchez]

Hi,

With human samples everything ran smoothly. Maybe if you want, send me the first ~ 50 rows of your MAF and I can try to run it with my script...

[reany]

Hi . I also encounter this issue. RunAbsolute works well locally,but when qsub to cluster,it returned the same error. Do you have any idea? thx.

Answer 1

[kmvaldez2]

I had a similar error in my run, it was caused by the mitochondrial 'chromosome' MT in my segmented_copy_ratios file not being an integer.

I understand that the original post was referring to the MAF, but check for this if you run into an issue with the seg file. I was going crazy trying to figure out what my error meant!

Answer 2

[kmvaldez2]

Also, none of the rows in the maf file can have a value of zero in BOTH t_alt_count AND t_ref_count.

Try something like the following to exclude them:

maf = maf[!(maf$t_alt_count==0 & maf$t_ref_count==0),]