With large sequencing projects, the generally practice it to have fastq files zipped to save space. If subread could read in zipped fastq files (.gz, .bz2, etc) as input this would be very useful.
