apytram FAQ

#apytram: Frequently Asked Questions

Speed optimization

Paired-end RNA-seq data run faster than single-end data.

To save time:

• not use -tmp option
  

• not use --keep_iterations
  

• not use --finish_all_iter
  

• not use the --plot and --plot_ali options
  

If you want to use apytram on several query files and you have several available threads, it is more efficient to minimize the number of threads by apytram job than to maximize the number of threads by job. Each job will be slower but at the end you will save time. This is due to the non linearity of the time saved by Trinity and Blast when the number of threads is increased.

Optimizing the accuracy

See folowing options:

• -e
• -mal
• -id
• -len
• -fid
• -fmal
• -flen

Following the program's progress

A job can take some minutes to several hours to complete. To know the progress of your job you can look into the log file (-log option or by default apytram.log).

You can look at the $OUTPUT_PREFIX.stats.pdf (--plot option) at the end the job, to have general information on the progress of your job. All values needed to create the plot in $OUTPUT_PREFFIX.stats.pdf are available in $OUTPUT_PREFFIX.stats.csv. The $OUTPUT_PREFFIX.stats.csv file can be only created using the --stats option.