Distinct alignment patterns

[Agathe Jouet]

Hi all,

I have a nucleotide alignment with 694 SNPs, from 30 individuals. All my sites are variable (in at least one individual). 

However, when I check the RaxML log file, I can see the following sentence:

Alignment has 65 distinct alignment patterns

I am not sure how RaxML gets to this number.

Could someone explain please?

Many thanks,

Agathe

[Alexey Kozlov]

Hi Agathe,

there is no contradiction here, since distinct alignment patterns are *unique* alignment *columns* (not to be confused
with constant/invariant columns).

Consider the following toy alignment:

AAA
AAC
TTC

It has 3 columns, all variable, but since columns 1 and 2 are identical, there are only 2 unique alignment patterns.

Best,
Alexey

On 15.03.2018 13:52, Agathe Jouet wrote:
> Hi all,
>
> I have a nucleotide alignment with 694 SNPs, from 30 individuals. All my sites are variable (in at least one individual).
> However, when I check the RaxML log file, I can see the following sentence:

> *Alignment has 65 distinct alignment patterns*

> I am not sure how RaxML gets to this number.
>
> Could someone explain please?
>
> Many thanks,
>
> Agathe
>

> --
> You received this message because you are subscribed to the Google Groups "raxml" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to raxml+un...@googlegroups.com
> <mailto:raxml+un...@googlegroups.com>.
> For more options, visit https://groups.google.com/d/optout.