I was running STAR on my local computer with i5 8 generation processor and 8 gigs of RAM. I was implementing a variant calling pipeline on simulated data of the ebola virus. I started with forward reads (.fq) and reverse reads (.fq) along with the reference genome of ebola(.fasta). I successfully used GATK to make reference genome index. While I was doing alignment with STAR (STAR --genomeDir ../genomedir/ --readFilesIn ../ebola_mut_reads1.fq ../ebola_mut_reads2.fq --runThreadN 4), I encountered an error.
Error is as follows
Dec 17 15:00:05 ..... started STAR run
Dec 17 15:00:05 ..... loading genome
Dec 17 15:00:06 ..... started mapping
Segmentation fault (core dumped)
Did you create the ebola index with the STAR aligner? You should have used something like
STAR --mode genomeGenerate --genomeDir /path/to/genomeDir --genomeFastaFiles /path/to/ebola.fasta ....
Yes, I did, I used it along with GATK and samtools as follows,
samtools faidx ebola_ref.fasta
gatk CreateSequenceDictionary --REFERENCE ebola_ref.fasta --OUTPUT=ebola_ref.dict
STAR --runMode genomeGenerate --genomeDir ../genomedir/ --genomeFastaFiles --genomeSAindexNbases 6 ebola_ref.fasta --runThreadN 4
Usually, STAR needs a lot of RAM to align and even more to build the index. Thus, having the STAR index files properly generated, you should be able to run the alignment.
Are there any clues in the log files from STAR, what went wrong? Have you tried it with R1 reads only?
I have used with R1 and R2 reads together as they are forward/reverse reads. Also, where I could find the log for STAR, I have installed it using 'make STAR'