What organism do you study and by using what software for making alignment? Full sequence you must obtain using such primers is small fragment of 18S-ITS1-5.8S-28(26)S small fragment. ITS1 is forvard; ITS4 - reverse primers. In this regin indels are not usual, but this region has a lot of multiple copy in genome (and this copy can be not totally identical), so one of efficient methods to eliminate this factor is DNA cloning. When you compare sequences keep in mind that one of them must be in reverse complement. Another factor which may cause some problems is fact that you use universal primers, which are workable with huge number of organisms. So, check your sequences by BLAST to be sure that is no contamination in your sequences.

I am guessing that these sequences are from a Sanger platform, is that correct? If so, then you are most likely having a drop in quality because you have a - possibly heterozygous - mix of DNA in your sample (e.g. with and without indel; or two indels of different length) . Because of this, the alignment will be shifted from the indel onward on some (but not all) reads. If the indel was the same in all DNA molecules of the sample, then the sequences should not show a quality drop after the indel.

With your existing data, cutting off the reads is basically the only possibility, as you cannot separate the mixed sequencing signals after the sequencing run. An alternative to cutting the sequences manually would be to adapt the alignment settings to align smaller parts of the reads rather than the whole reads, and then fade out the parts that don't match - but in the end this will lead to the same result as cutting off the reads manually. The downside to cutting off the sequences, no matter how you do it, is that you do not get exact information on the type/size of the indel.

To really solve this problem I think you will have to change you sequencing approach:

Most NGS platforms (Illumina, 454, etc.) would solve this problem by paired-end sequencing (sequencing one molecule from both sides). This would be my first suggestion if you have the possibiliy to access an NGS platform.

If you do not have this option, another option would be the much more time-consuming method of cloning and sequencing. By doing this, you will clonally amplify a single DNA molecule in bacteria, and then sequence it. This way you will not have a mix of DNA in your sample. However, you will have to sequence a whole bunch of clones to judge the composition of your sample (so you will need to sequence 10 clones at the very very least).

In short: I suggest to try a different sequencing approach if you have the time and budget - especially if you need exact info on the indel. If not, cut off the sequences manually, do the alignment and move on to the next project  :-)

Hope this helps - Good luck!

Hi!

Thank you very much for you help! 

To be honest i had already considered those options, and regarding the alternative sequencing methods, the time and cost that they have are unfortunately not suited for this work... too costy :) Anyway, i did confirm the organisms using Blasts, and i am using ITS4B which is for basidiomycetes, which reduces the contamination probability. 

In fact i also think its something to do with different sequences within each organism that is leading to this, and in order to attain a good sequence i am really considering cutting out the bad portions of each of the directions, which will provide me with at least a good read for the first half (in one direction) and the second half (in the other direction) so the consensus will more or less be robust. I am also considering aligning the obtained consensus itself with one of the sequences in which i didn't have this problem, and generate a stronger, more robust consensus sequence for each population, assuming that they are all alike within the population (since i already know the organism and am just wanting to assert different populations from different places).

I wanted to know if there were any alternative methods to help me with this, or any recommended approach, that would be more robust, since i am afraid it might bias the results... this is a tough decision =)

In any way, thank you once more for your support!

Some basidiomicetes really have inrons in this region. So to solve this problen you may firstly use DNA cloning to avoid of presence of different copies. In addition you also may use more specific primers.