STAR can't genereate genome Index.

Question

[e183021g]

Hi Everyone .

I can't make Index file by using the latest STAR as follows on Linux. My PC has 16 cores and 64GB RAM. There was no ERROR message.

~/Documents/expression/tools$ ~/Documents/expression/tools/STAR-2.7.10b/bin/Linux_x86_64_static/STAR --runThreadN 10 --runMode genomeGenrate --genomeDir ~/Documents/expression/ref/STAR_reference --genomeFastaFiles ~/Documents/expression/ref/Homo_sapiens.GRCh38.dna.primary_assembly.fa --sjdbGTFfile ~/Documents/expression/ref/Homo_sapiens.GRCh38.95.gtf --sjdbOverhang ReadLength-1

Nov 04 13:25:00 ..... started STAR run

>~/Documents/expression/tools$ echo $?

1

I am very new about STAR and bioinformatics . Would you tell me anything about how to treat with this error?

Shintaro

Answer 1

[iraun]

Hi! And welcome Biostars! STAR is a very RAM needy aligner, but I believe that 64GB should suffice to map to the human genome. Have you tried to run the command using just the sequence of one chromosome instead of the whole genome to see if it runs? Alternatively, I think you can find some genome indexes online as in here. But it is generally a good practice to create the index before running the alignment step. Btw, you have a typo in the command, it is genomeGenerate. And I guess the variable ReadLength contains the actual length of the reads? Also, do not expect to have the results fast, it will take at least several hours to generate the indexes, so maybe you just have to wait a bit longer :)

[e183021g]

Thank you for your great advice!! I altered genomeGenrate to genomeGenerate and STAR started to make genome Index which I want to get. As you said , It was a very good practice for me to create the index. Thank you so much for your help and read my poor English!!