Question: (Closed) Encountering Paml Error When Using Lazarus Package
gravatar for Ncemami
8.5 years ago by
Ncemami10 wrote:

I am trying to use Lazarus, which is an ancestral reconstruction program using PAML, on a set of protein sequences.
The Lazarus portion of things seems to be working just fine, I invoked with pretty much standard parameters:

  LAZARUS for PAML version June.28.2012
  questions? contact Victor Hanson-Smith

Settings for this run:

  D    datatype                 => amino acid
  A    sequence alignment file  => fp.fasta
            => O.K. I found 305 taxa.
  T    newick tree file         => fp.newick
            => O.K. I found 1 tree(s).
  P    path to folder w/ models => /home/nimaemami/Lazarus/Lazarus/paml/dat
            => O.K. I found 17 models named with *.dat or *.rates
  M    evolutionary model       => JTT
  B    branch lengths           => find ML estimate
  V    rate of evolution        => use 4 gamma-distributed categories
  O    output to this location  => /home/nimaemami/FPs
  I    place ancestral gaps     => No, I'll manually do it later.
  U    outgroup taxa            => ['259','167']

Are these settings correct? (type Y or letter to change option)Y
Invoking Lazarus with this command:
lazarus_batch --codeml --outputdir /home/nimaemami/FPs --alignment fp.fasta --tree fp.newick --model /home/nimaemami/Lazarus/Lazarus/paml/dat/jones.dat --branch_lengths estimate --asrv 4 --outgroup [['259','167']]
2.7.1+ (r271:86832, Apr 11 2011, 18:13:53)
[GCC 4.5.2]
--> Importing your files. . .
I found your alignment file: fp.fasta appears to contain 305 taxon.
I found your tree file: fp.newick appears to contain 1 trees.
I found your evolutionary model at: /home/nimaemami/Lazarus/Lazarus/paml/dat/jones.dat
--> Starting PAML jobs. . .

Once it starts calling PAML, though, it is encountering the following error:

Error: initial value alpha <= 0 for fix_alpha = 0.
Hmmm, The PAML job on tree 1 did not produce a file named 'out.paml'.  I was expecting to find this file at /home/nimaemami/FPs/tree1/pamlWorkspace/out.paml

Has anybody encountered this before? I don't think the problem is the Lazarus model I'm using, since I tried another model and got the same result. It seems like there is some internal PAML parameter that I didn't touch that is breaking things. Any insight into how to get around this would be greatly appreciated.


protein paml • 2.6k views
ADD COMMENTlink written 8.5 years ago by Ncemami10

I think I ran into this in the past. I seem to recall that the problem was that the trees could not be written to the specified location. Can you run paml from commandline? does that work?

ADD REPLYlink written 8.5 years ago by Whetting1.5k

Hi Whetting -- I have baseml and codeml binaries installed, which is supposedly sufficient for running Lazarus.

ADD REPLYlink written 8.5 years ago by Ncemami10

Hi, if you try running codeml from terminal what happens?

ADD REPLYlink written 8.5 years ago by Whetting1.5k

I recently encountered this problem myself. Running codeml using the automatically generated codeml.ctl file threw the same error message (Error: initial value alpha <= 0 for fix_alpha = 0). There are two solutions to this problem:

You can set fix_alpha = 1 by specifying "use fixed rate across sites" in Lazarus, but this could affect your reconstructions.

Alternatively, you can download the updated version which can use ML estimates to provide a non-zero alpha value. The updated version can be found at (just clone it and have fun!)

ADD REPLYlink written 8.3 years ago by ericm.zinn0

Hi Nima, this is Victor Hanson-Smith (the author of Lazarus). The error you encountered is fixed in the latest version of Lazarus. []. If you are still having problems, please email me at

ADD REPLYlink written 7.0 years ago by victorhansonsmith0
Please log in to add an answer.
The thread is closed. No new answers may be added.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1309 users visited in the last hour