Thanks. CAP3 is perfect for this.
Hi. I have hundreds of paired-end sequences produced by Sanger (length ~800 bp) that I would like to assemble by pair. The overlapping region can vary from 20 bp until 150 bp. The difficulty is that some "N" in the overlapping area can occur and one strain have to be reverse transcribed. What tool should I use to address my problem? Thanks a lot.
1 answer
If I understand your question, you want to assemble two sequences at a time, one forward strand, and the other reverse strand, the two sequences meet in the middle. Pretty much any overlap-based assembler can do the trick. However, since there are only two sequences, you need to decide how to call consensus bases when the overlapping region of the two sequences disagree.
Important parameters are max allowed hangs, identity cutoff, and minimum overlap length. Be aware that these parameters affect the outcome, even though you are likely going to use the defaults.
overlap hang
>>>>>>>>>>>>>>>>>>>>>>>>>>>>> Forward
|||||||
<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Reverse
For command-line programs, use EMBOSS merger or CAP3. I am not familiar with many GUI options, but have used Sequencher in the past to assemble small number of reads.
If there are substantial number of bad bases of N's towards the end of each sequence (which will prevent two sequences from merging), consider trimming based on quality.
Log in to answer this question.