This is more of an efficiency question - the BWA 'sampe' command provides an option to 'pre-load' the index into memory. If this option is not selected, BWA has to reload the index into memory every 262144 reads. It seems inefficient, for instance if you had 262145 reads, you would reload the index just to map one more read.
I'm just wondering what 'base-space' reads are since it's only available for those (I'm assuming Illumina, as SOLiD reads are considered colour space?), and why this option wouldn't be a default since it really seems to be much more efficient?
Also, if someone somehow knows this, I'm wondering why this option wouldn't be available for colour space reads.
I'm just wondering what 'base-space' reads are since it's only available for those (I'm assuming Illumina, as SOLiD reads are considered colour space?), and why this option wouldn't be a default since it really seems to be much more efficient?
Also, if someone somehow knows this, I'm wondering why this option wouldn't be available for colour space reads.
Comment