Segmentation fault (core dumped) in the alignment step

[Vasilis Lenis]

Dear Alex,

I am trying to use STAR in order to map some paired end reads in my reference bacteria genome (Synechocystis,)

I have created the index without any trouble but when I am trying to run the alignment step I'm getting a segmentation fault without any notification (the log file doesn't include anything helpful for the problem). I'm using the last STAR version (020201) without any "fancy" parameters:

/STAR-master/bin/Linux_x86_64_static/STAR --genomeDir /data/vlenis/STARindex --readFilesIn /trimmed/3a_ATGTCA_L001_R1_001.p.fastq /trimmed/3a_ATGTCA_L001_R2_001.p.fastq --outFileNamePrefix /alignment_STAR/3a_ATGTCA_L001 --outFilterMultimapNmax 1 --outReadsUnmapped Fastx --outSAMtype BAM SortedByCoordinate --twopassMode Basic --runThreadN 1

I have also tried to run it without any extra parameter (STAR --readFilesIn reads_R1.fq reads_R2.fq --readFilesIn mapped) but the problem still remains.

Thank you very much in advance,

Vasilis.

[Alexander Dobin]

Hi Vasilis,

if you have a small genome, you would need to scale down --genomeSAindexNbases as log2(GenomeLength)/2 - 1 (but no less than 14). This should eliminate the seg-fault at mapping.

Cheers

Alex

[David Vuono]

I am experiencing the same Segmentation fault (core dump) error running in alignReads run mode. We have 4Mb bacterial genome. We've tried to rescale the --genomeSAindexNbases option but this doesn't work either. It seem weird that you would say (no less than 14) when 14 is also the default. I know this script has run on even smaller genomes (i.e., Nitrosopumilus maratimus: ~1.5Mb) (authors of that paper don't like to respond to emails). These are our scripts (STAR --version STAR_2.5.3a):

STAR --runThreadN 20 --runMode genomeGenerate --genomeDir path/to/genome --genomeFastaFiles path/to/fasta.fa --sjdbGTFfile path/to/annotation.gtf --genomeSAindexNbases 10 --sjdbOverhang 100 &

STAR --runThreadN 20 --runMode alignReads --genomeDir path/to/indexed/genome_dir --readFilesIn path/to/forward/R1_read.fq path/to/reverse/R2_read.fq --outSAMtype BAM SortedByCoordinate --outFileNamePrefix EE1

[Alexander Dobin]

Hi David,

sorry for the confusion, this was a typo, it should read "no more than 14".

Please try to further reduce the --genomeSAindexNbases to say, 6 or even 4.

Cheers

Alex