Input files reference and features have incompatible contigs

Hi
I am very new to GATK. i read the paper Curr Protoc Bioinformatics. ; 11(1110): 11.10.1–11.10.33. doi:10.1002/0471250953.bi1110s43 and best practices guide. i want to run whole exom analysis to find high confidence SNP and indels .I am stuck at the BQSR analysis step ( BaseRecalibrator). it shows the error message
Input files reference and features have incompatible contigs: No overlapping contigs found.

i did following steps
1. Download the genome file hg38.fa.gz from UCSC genome.
2.Indexing the reference genome : bwa index hg38.fa
3.Create fasta file index : samtools faidx hg38.fa
5. Create sequence dictionary by java -jar picard.jar CreateSequenceDictionary REFERENCE=hg38.fa OUTPUT=hg38.dict
6. mapping the data to reference : bwa mem -R '@RG\tID:group1\tSM:sample1\tPL:illumina\tLB:lib1\tPU:unit1' -p hg38.fa R1.fastq R2.fastq > aligned_reads.sam
7. sort my align reads: java -jar picard.jar SortSam INPUT=aligned_reads.sam OUTPUT=sorted_reads.bam SORT_ORDER=coordinate
8. marking duplicates : java -jar picard.jar MarkDuplicates INPUT=sorted_reads.bam OUTPUT=dedup_reads.bam METRICS_FILE=metrics.txt
9. index my markduplicate file : java -jar picard.jar BuildBamIndex INPUT=dedup_reads.bam
10. Download dbSNP from this ftp://ftp.ncbi.nih.gov/snp/organisms/human_9606_b149_GRCh38p7/VCF/00-All.vcf.gz
11. Download indels from ftp://ftp.1000genomes.ebi.ac.uk/vol1/ftp/technical/reference/GRCh38_reference_genome/other_mapping_resources/Mills_and_1000G_gold_standard.indels.b38.primary_assembly.vcf.gz
12. Run the BaseRecalibrator : java -jar gatk-package-4.0.0.0-local.jar BaseRecalibrator -R hg38.fa -I dedup_reads.bam --known-sites SNP_00-All_GRCh38.vcf --known-sites Mills_and_1000G_gold_standard_indels.vcf -O recal_data.table
it shows the error message to index the SNP_00-All_GRCh38.vcf and indel file
13.i index the file by running this command : java -jar gatk-package-4.0.0.0-local.jar IndexFeatureFile -F SNP_00-All_GRCh38.vcf and java -jar gatk-package-4.0.0.0-local.jar IndexFeatureFile -F Mills_and_1000G_gold_standard_indels.vcf

  1. After index, again i run the BaseRecalibrator : java -jar gatk-package-4.0.0.0-local.jar BaseRecalibrator -R hg38.fa -I dedup_reads.bam --known-sites SNP_00-All_GRCh38.vcf --known-sites Mills_and_1000G_gold_standard_indels.vcf -O recal_data.table

it shows the error " A USER ERROR has occurred: Input files reference and features have incompatible contigs: No overlapping contigs found."

Please help me to solve the problem

Best Answer

Answers

  • Hi @EADG

    Thank you. I will try this

  • rzliurzliu Member

    Hello @EADG

    I have a similar error when change the resource bundle from b37 to hg19. For b37, the Mutect2.gnomad is set to "af-only-gnomad.raw.sites.vcf" and Mutect2.variants_for_contamination is set to "small_exac_common_3.vcf." What's the correct files from hg19 resource bundle for these two values?

    Thank you very much for the help!

  • SheilaSheila Broad InstituteMember, Broadie, Moderator admin

    @rzliu
    Hi,

    You will need to do a liftover for those b37 files to hg19. You can use LiftoverVcf.

    -Sheila

  • qibaoqibao Member
    Hi,
    I met a similar problem. when I tried to run the following code:

    gatk BaseRecalibrator \
    -R ./RefGenome/GCF_000001405.38_GRCh38.p12_genomic.fa \
    -I ./Results/SRR2481145_dedup_split.bam \
    --known-sites ./resources/bundle/hg38/1000G_phase1.snps.high_confidence.hg38.vcf \
    --known-sites ./resources/bundle/hg38/Mills_and_1000G_gold_standard.indels.hg38.vcf \
    --known-sites ./resources/bundle/hg38/dbsnp_146.hg38.vcf \
    -O ./Results/SRR2481145_recal_data.table

    And I got the following error:

    A USER ERROR has occurred: Input files reference and features has incompatible contigs : No overlapping contigs found

    reference contigs = [NC_00001.11, NT_187361.1 ...
    features contigs = [chr1, chr2, chr3, ...

    My question is shall I use the reference genome in the bundle (Homo_sapiens_assembly38.fasta) ?
    And a further question is should the reference genome used here be the same one used in the STAR mapping step ?
  • AdelaideRAdelaideR Unconfirmed, Member, Broadie, Moderator admin

    Hi @qibao

    Yes, the contigs do have to be named the same for the program to work. So, using the reference genome in the bundle will provide the contigs with the "chr1, chr2, chr3" naming.

    If you want to combine data sets, using the same reference genome for mapping will facilitate this.

Sign In or Register to comment.