Write a report that characterizes the destination of repeat reads. It would
designate some reads as repetitve and then show what portions of them
were output as singletons, degenerates, and placed contigs.
This analysis should recognize repetitive reads based on their k-mer
content. For instance, it might identify a 20-mer as frequent if it occurs at
least 100 times in the fragment set. Then it might designate a read as
repetitive if it has frequent 20-mers over 2/3 of its clear range.
This analysis might use a clustering algorithm. For instance, it might
report that 100K reads have a certain cluster of frequent k-mers, and 90K
of those reads ended up as singletons.
Log in to post a comment.
