RAxML error: Problem reading number of species and sites

[bfp7]

Hi ,
I am trying to make a tree with RAxML that I will then use with placer.

When I run the raxml command I get the error: Problem reading number of species and sites.

I googled the error and used the sed command to remove spaces from the header of the aligned nucleotide file. I used muscle to align my sequences. This is what my header looks like after removing spaces:

>ENA|CAJ48085|CAJ48085.1Bordetellaavium197Nbiodegradativeargininedecarboxylase
ATGAAATTTCGCTTCCCCATTTTCATCATCGACGAAGACTTCCGTTCCGAGAACGCCTCG

The raxml manual indicates that identical sequences are a problem. This is from the "Alignment Error Checking" section of the manual:

2. Identical Sequence(s) that have di erent names but are exactly identical. This mostly happens when you excluded some hard-to-align alignment regions from your alignment and does not make sense to use.

My sequences have a high percent identity. Uclust percent identity output for my sequences looks like 100%, 100%, 99.9%, 99.9%, 99.8%, 99.8% and some of the genus/species names are the same too. Is this the source of the error? Or maybe it is something else.

Is there a tree building software that I can use with placer that can handle sequences where some have 100% identity ???

Thanks!! Sorry this is so long.

[cliffbeall]

It looks like your sequence is in fasta format, I think you need to get them into Phylip format for RAxML. I do that with Mesquite on the Mac, biopython alignIO will also work.

[bfp7]

Hi cliffbeall,
Thx for the reply. I installed RAxML 8.2 and it accepts fasta files. I am no longer getting that error message when I run RAxML. But now when I run pplacer I get an error message:

Running pplacer v1.1.alpha17-6-g5cecf99 analysis on Q2KYQ0_BORA1.fasta...
Didn't find any reference sequences in given alignment file. Using supplied reference alignment.
Warning: using a statistics file directly is now deprecated. We suggest using a reference package. If you already are, then please use the latest version of taxtastic.
WARNING: your stats file is from RAxML 8.2.8; RAxML has been tested with the following versions: 7.0.4; 7.2.3; 7.2.5; 7.2.6; 7.2.7
I'm going to try parsing as if this was version 7.2.3Problem parsing info or stats fileRAxML_info.10_Q2KYQ0
Uncaught exception: Parse_stats.Stats_parsing_error("too many partitions. Only one is allowed.")

I did include a reference alignment. It is the same alignment that RAxML used to make the RAxML_result file. Maybe pplacer does not accept the newer version of RAxML???

I am at a loss. I am also a novice. Any additional help would be appreciated!
Thx,
Brian