Hello Andrii,

From the sound of the question the sequences seem to be aligned already and do not contain indels but instead just extra sequence. There are a lot of ways to handle this and sometimes it is better to maintain extra data rather than trim it off and other times it it better to trim. It also depends on what program you are using and how it treats missing data. The question I posted a link for gives a lot of good suggestions on what can be done with dealing with missing data. Also below is another question that has a good discussion as well.

https://www.researchgate.net/post/How_do_I_build_a_phylogeny_with_missing_data_and_incomplete_taxon_coverage

I also think the publication below gives good background on this issues and empirically test the effects of missing data (although it is more geared to species tree analysis which involves multiple loci typically generated by NGS, but it is still relevant to this discussion).

https://www.researchgate.net/publication/284273579_The_impact_of_missing_data_on_species_tree_estimation

Best Regards

Article The Impact of Missing Data on Species Tree Estimation

Hello Adam,

Here is a similar question that you may find useful.

https://www.researchgate.net/post/How_do_I_choose_between_loosing_information_and_including_too_many_gaps_when_building_phylogenetic_tree/1

Best Regards

First of all, you'll need to have your sequences aligned. After this, you can trim your sequences because there are no good mathematical algorithms to involve indels in phylogenetic analysis.

Hello Andrii,

From the sound of the question the sequences seem to be aligned already and do not contain indels but instead just extra sequence. There are a lot of ways to handle this and sometimes it is better to maintain extra data rather than trim it off and other times it it better to trim. It also depends on what program you are using and how it treats missing data. The question I posted a link for gives a lot of good suggestions on what can be done with dealing with missing data. Also below is another question that has a good discussion as well.

https://www.researchgate.net/post/How_do_I_build_a_phylogeny_with_missing_data_and_incomplete_taxon_coverage

I also think the publication below gives good background on this issues and empirically test the effects of missing data (although it is more geared to species tree analysis which involves multiple loci typically generated by NGS, but it is still relevant to this discussion).

https://www.researchgate.net/publication/284273579_The_impact_of_missing_data_on_species_tree_estimation

Best Regards

Article The Impact of Missing Data on Species Tree Estimation

Hello,

Two things should be taken in your consideration at first;all sequences should include region(s) that you study particularly if you have variable regions in your sequences, and if you will obtain deduced amino acids or not. However, it is more accurate to trim the sequence if it contains regions of study and the length difference was small.

There seems to be no strict limits on the length of the genes. Indeed, the gene rpmJ encoding the ribosomal protein L36 is only 111 bp long in most bacteria, whereas the gene for B. subtilis polyketide synthase PksK is 13,343 bp long. In practice, mRNAs shorter than 30 codons are poorly translated, so protein-coding genes in prokaryotes are usually at least 100 bases in length. In prokaryotic genome-sequencing projects, open reading frames (ORFs) shorter than 100 bases are rarely taken into consideration, which does not seem to result in substantial under-prediction. In contrast, in multicellular eukaryotes, most genes are interrupted by introns. The mean length of an exon is ~50 codons, but some exons are much shorter; many of the introns are extremely long, resulting in genes occupying up to several megabases of genomic DNA. This makes prediction of eukaryotic genes a far more complex (and still unsolved) problem than prediction of prokaryotic genes.

Hi Adam,

I agree with previous comments about sequences being aligned prior to any phylogenetic reconstruction. Once sequences have been aligned, by definition, all sequences will have the same length thanks to the use of gaps.

As pointed in one of the RG links, you may trim (or not) the alignment to improve the general accuracy of the downstream tree. Trimming or not will depend on 1) the general amount of variable columns - columns containing gaps - and 2) the method used to reconstruct the phylogenetic tree because there are methods less sensitive e.g. bayesian and maximum-likelihood to the presence of gaps than others e.g. NJ and UPGMA.

Hope it helps,

S

Hi Everyone,

Thanks for your helpful responses! To confirm, sequences are aligned, my question was regarding 'extra' flanking sequence.

Thanks again,

Adam

Th extra length occurring at the beginning and end may simply be an artifact of the performance of the primers in different taxa or the quality of the each batch of DNA. 

If only one sequence is longer than do trim it. Or if more than one sequence and the extra stuff appears noisy (little apparent homology, then also trim it. If the extra sequence appears phylogenetically informative for those taxa that have longer sequences then you may want to retain it. ML analysis will ignore the missing data but will still contribute to the overall likelihood (the sum of the site likelihoods) .Any phylogenetic information at those extra sites should help resolve those sequences but not at he expense of the shorter sequences provide.

Dear Adam 

Yes, you are right, All the sequences must be align first then use for phylogenetic analysis.