Cannot milk it?
Once you detect the bad quality letters in your reads, a good strategy is not to filter out entire reads (as we do in “Read Filtration by Quality”) but trim low-quality bases from both ends of your reads. it helps to save a large amount of information.
Bad quality bases can be easily trimmed out using certain threshold (defined by quality plot similar to what we did in “Base Quality Distribution”) There is a lot of trimming tools, you can try one of following:
Given: FASTQ file, quality cut-off value
Return: FASTQ file trimmed from the both ends (removed leading and trailing bases with quality lower than
20 @Rosalind_0049 GCAGAGACCAGTAGATGTGTTTGCGGACGGTCGGGCTCCATGTGACACAG + FD@@;C<AI?4BA:=>C<G=:AE=><A??>764A8B797@A:58:527+, @Rosalind_0049 AATGGGGGGGGGAGACAAAATACGGCTAAGGCAGGGGTCCTTGATGTCAT + 1<<65:793967<4:92568-34:.>1;2752)24')*15;1,.3*3+*! @Rosalind_0049 ACCCCATACGGCGAGCGTCAGCATCTGATATCCTCTTTCAATCCTAGCTA + B:EI>JDB5=>DA?E6B@@CA?C;=;@@C:6D:3=@49;@87;::;;?8+
@Rosalind_0049 GCAGAGACCAGTAGATGTGTTTGCGGACGGTCGGGCTCCATGTGACAC + FD@@;C<AI?4BA:=>C<G=:AE=><A??>764A8B797@A:58:527 @Rosalind_0049 ATGGGGGGGGGAGACAAAATACGGCTAAGGCAGGGGTCCT + <<65:793967<4:92568-34:.>1;2752)24')*15; @Rosalind_0049 ACCCCATACGGCGAGCGTCAGCATCTGATATCCTCTTTCAATCCTAGCT + B:EI>JDB5=>DA?E6B@@CA?C;=;@@C:6D:3=@49;@87;::;;?8