Hi,
I am trying to map solid reads using bfast, and have several questions on using it, any help will be highly appreciated!
1) Is it always necessary to partition reads into bins? In a post, Nils suggested to do that, but I only have ~20M reads, and a cluster with large-memory nodes, is it still necessary to do this step?
2) In postprocess step, there are several filter options, like "-a 2 outputs only
reads that have been aligned uniquely", and "-a 3 chooses uniquely the alignment with the best score". Could someone explain the difference between these two options? It seems identical to me.
Best,
Tree
I am trying to map solid reads using bfast, and have several questions on using it, any help will be highly appreciated!
1) Is it always necessary to partition reads into bins? In a post, Nils suggested to do that, but I only have ~20M reads, and a cluster with large-memory nodes, is it still necessary to do this step?
2) In postprocess step, there are several filter options, like "-a 2 outputs only
reads that have been aligned uniquely", and "-a 3 chooses uniquely the alignment with the best score". Could someone explain the difference between these two options? It seems identical to me.
Best,
Tree