search

JiekaiLab / scTE

MIT License
87 stars 27 forks source link

scTE does not finish, stuck on Fetching from annotation file #2

Closed HectorRDB closed 3 years ago

HectorRDB commented 3 years ago

Hi, Thanks for this software, it's great and intuitive to use. I was trying to analyze the data from Deng et al, using the first sample as an example. I build the mm10 genome with 

scTE_build -g mm10 -o /scratch/reference/scTE/mm10

Then I tried to analyze one sample. The fastq file were obtained from GEO and aligned using STAR

STAR --genomeDir /scratch/Databases/Mm10/STAR_genome/ \
  --readFilesIn /scratch/singleCell/Mouse/GSE45719_Deng_2014/Fastq//SRR805173.fastq.gz \
  --readFilesCommand gunzip -c --outFileNamePrefix /scratch/Deng/bam/SRR805173/SRR805173 \
   --outSAMtype BAM SortedByCoordinate --outSAMattributes XS --runThreadN 16 \
  --outFilterMultimapNmax 100

Then I tried to run scTE

scTE -i bam/SRR805173/SRR805173Aligned.sortedByCoord.out.bam -o counts_ -g mm10 \ 
  -x /scratch/reference/scTE/mm10.exclusive.idx -p 16    --expect-cells 300 -CB False -UMI False

However, it is blocked for hours now after printing INFO: fetching from the annotation index 2021-03-17 16:05:27

Do you have any idea what could cause this? Best

jphe commented 3 years ago

That message looks like an out of memory error, scTE takes ~10Gb memeroy for each thread, you can change the -p option to be smaller.

HectorRDB commented 3 years ago

Thank you that indeed solved the issue. It would be nice to have this info in the Read Me or the help file somewhere to help calibrate the number of cores one uses.