Picard error: Sequences at index 0 don't match

Question

[zachariasawaged]

I am having an issue with Picard when I try to use the CollectAlignmentSummaryMetrics tool.

So I aligned my Fastq into a BAM using Star and used a pre-made Star index from this directory: http://labshare.cshl.edu/shares/gingeraslab/www.data/dobin/STAR/STARgenomes/UCSC/hg19_inclHap_noAnnotations_SAsparseD3/

And then I am using these chromosomes.fa files as my reference for Picard: http://labshare.cshl.edu/shares/gingeraslab/www-data/dobin/STAR/STARgenomes/UCSC/hg19_noAnnotations/FASTA/

I used this code to put the chr.fa files together into one fasta file and make a dict:

tar -zxvf chromFa.tar.gz
cat chr*.fa > hg19.fa
module load gatk
gatk CreateSequenceDictionary -R hg19.fa

Then when I run Picard, here is the error:

Exception in thread "main" htsjdk.samtools.util.SequenceUtil$SequenceListsDifferException: Sequences at index 0 don't match: 0/249250621/chr1 0/135534747/chr10/M5=988c28e000e84c26d552359af1ea2e1d/UR=file:/gpfs/scratch/sawagz01/hg19.fa

Any advice? Is there an issue with using two HG19 sequences from two different directories?

[Pierre Lindenbaum]

then when I run Picard

which tool ? which parameters ??

most probably you merged the fasta sequence

cat chr*.fa > hg19.fa

with an order that is not the same as the one used by a bam or a vcf file.

[zachariasawaged]

Hey Pierre - I mentioned it in my fist line: CollectAlignmentSummaryMetrics tool. How do I adjust the order?

Answer 1

[swbarnes2]

Are you completely sure that your pre-made index was made using the chromosome.fa that you are giving to Picard?

[zachariasawaged]

That's my main concern. They were both from UCSC but not from the same file. I downloaded the index pre-made, and used a fa file that was in a different directory than my index.

I ended up realigning to an fa file that was in the same directory, but that gave me a new error: Sequence dictionaries are not the same size (84, 93)