Hi, everyone
I was analyzing a vriome dataset these days. I aligned about 30 milions reads with the length of 150 bp to the NT database with BLASTn (I used botiew2 as well for confirming). Around 10 milions reads were mapped to the same gene on the basis of e-value (1e-10), but with low query cover (40~50bp/150bp). Though the result can be accepted based on the e-value, someone suggested that these alignments should be removed because of the low query cover. Such abnormal alignments (plenty of reads from different samples are exactly mapped to the same gene with low query cover 40-50bp) occur in many samples.
The dataset have been preprocessed by the sequencing company and they just sent me the clean reads. I want to ask that whether or not they should be seen as contamination and be removed under a certain threshold (I have not seen a paper use query cover as the exclusion criteria, in virome area. All use e-value. ), and why such reads appear? If they are contamination, how should I address this problem?
Thank you very much for your kind help!
Best wishes, Yanshan
