Partition Homogeneity Test on PAUP

[Jeremiah Sia]

Hi, 

I ran partition homogeneity test on PAUP for around 6 hours but still the task is not completing so I terminated it. 

Would like to ask why does it take such a long time? is it normal? or i have incorrectly set the program?

Thank you.

Jeremiah

[Mark Miller]

Hi Jeremiah,

That seems like a long time with no progress. I am looking into the issue. An alternative code is MPBoot, which may or may not meet your needs if you want to try an alternative that is supposed to be faster.

Sincerely,

Mark Miller. 

[Jeremiah Sia]

Thank you Mark for your reply.

Will look into MPBoot.

Regards,

Jeremiah

--
You received this message because you are subscribed to a topic in the Google Groups "CIPRES Science Gateway Users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/cipres-science-gateway-users/W6iWroGFxME/unsubscribe.
To unsubscribe from this group and all its topics, send an email to cipres-science-gatew...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cipres-science-gateway-users/a7160cf2-b725-45a3-965a-ecbeac897d82n%40googlegroups.com.

[Miller, Mark]

Great.

My colleague reminded me that PAUP reports no progress as it runs, so it may be ok.

I am running a test now.

 

Mark

[Jeremiah Sia]

Hi Mark,

Noted. Would like to know if it is due to any mistakes from my part. 

Thanks 

Jeremiah 

[Miller, Mark]

Hi Jeremiah,

 

We did some experimentation with your data set, and here is what we found.

If we change the set increase from auto to no, the results appear very quickly.

As I understand it, the auto option allows PAUP to increase the number of trees until it has converged.

I copied the results in stdout.txt below. I am by no means a PAUP expert, and I would like to know wht kind of output you are expecting.

The output I see is below.

 

With no change to maxtrees, I get results like this:

 

Logging output to file "/expanse/projects/ngbt/backend/expanse_test_workspace/NGBW-JOB-PAUP_XSEDE-85E71402FFFF4AB7B2780AF6AC45BCEE/infile.log".

 

Partition-homogeneity test with heuristic search:

  Character partition = genes

  Starting seed = generated automatically

  Number of replicates = 100

  Optimality criterion = parsimony

    Character-status summary:

      Of 5338 total characters:

        All characters are of type 'unord'

        All characters have equal weight

        4709 characters are constant (proportion = 0.882166)

        14 variable characters are parsimony-uninformative

        Number of parsimony-informative characters = 615

    Gaps are treated as "missing"

  Starting tree(s) obtained via stepwise addition

    Addition sequence: random

    Number of replicates = 10

    Starting seed = generated automatically

    Number of trees held at each step = 1

  Branch-swapping algorithm: tree-bisection-reconnection (TBR) with reconnection limit = 8

    Steepest descent option not in effect

  'Maxtrees' setting = 100 (will not be increased)

  Branches collapsed (creating polytomies) if maximum branch length is zero

  'MulTrees' option in effect

  No topological constraints in effect

  Trees are unrooted

 

  100 partition-homogeneity test replicates completed

  Time used = 00:03:13 (CPU time = 00:03:12.2)

 

Results of partition-homogeneity test:

 

        Sum of       Number of

  tree lengths       replicates

  -----------------------------

           757*           1

           758            3

           759            2

           760            3

           761           10

           762            6

           763            8

           764           13

           765           16

           766           17

           767            9

           768            7

           769            4

           770            1

              * = sum of lengths for original partition

 

   P value = 1 - (99/100) = 0.010000

 

I can increase set maxtrees=10000, and the job will still run.

The results look like this:

  100 partition-homogeneity test replicates completed

  Time used = 01:52:13 (CPU time = 01:51:41.4)

 

Results of partition-homogeneity test:

 

        Sum of       Number of

  tree lengths       replicates

  -----------------------------

           757*           2

           759            2

           761            4

           762           12

           763            8

           764           16

           765           13

           766           13

           767           14

           768           10

           769            3

           770            3

              * = sum of lengths for original partition

 

   P value = 1 - (98/100) = 0.020000

 

Processing of input file "infile.nex" completed.

 

paup>

 

paup>lset nthreads=1;

 

paup>

 

paup>set autoclose=yes;

 

paup>quit;

paup>

--
You received this message because you are subscribed to the Google Groups "CIPRES Science Gateway Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cipres-science-gatew...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/cipres-science-gateway-users/CABDwMxBqG3UZY6Eh%3D9pY4Jd8qL1mcjHgPvYUw8JpVG3Pwr_j_g%40mail.gmail.com.

[Jeremiah Sia]

Hi Mark, 

Thank you for your email. I am actually not an expert on PAUP as well. I am expecting a p-value (just as the output you have shown me) that would allow me to decide whether the partitions are congruent enough to be concatenated. Typically a p-value less than 0.05 is good for me. I have never managed to obtain any output from the PAUP 4.0a169 software thus far. i had to manually increase the MaxTrees from time to time, whenever the system required it, yet the process will just come to an end without any output. When I turn on the auto set increase, then it won't end. So I turned to CIPRES. 

The output looks promising, I will try to adjust the MaxTrees number to see if it will affect the results. 

Thanks a lot. 

Regards,

Jeremiah

[Jeremiah Sia]

Hi Mark and team,

I have managed to figure out what went wrong, According to a paper (https://onlinelibrary.wiley.com/doi/full/10.1111/j.1365-313X.2005.02611.x), I have left out 1 line of command "nchuck=2 chuckScore=5" to save no more than ‘2’ trees ≥ score ‘5’ for each replicate. I managed to run it and get the results using the PAUP software while having the set Maxtrees remain in auto. The output as below:

Data matrix has 28 taxa, 5338 characters
Valid character-state symbols: ACGT
Missing data identified by '?'
Gaps identified by '-'
"Equate" macros in effect:
   R,r ==> {AG}
   Y,y ==> {CT}
   M,m ==> {AC}
   K,k ==> {GT}
   S,s ==> {CG}
   W,w ==> {AT}
   H,h ==> {ACT}
   B,b ==> {CGT}
   V,v ==> {ACG}
   D,d ==> {AGT}
   N,n ==> {ACGT}

Partition-homogeneity test with heuristic search:
  Character partition = genes
  Starting seed = generated automatically

  Number of replicates = 1000

  Optimality criterion = parsimony
    Character-status summary:
      Of 5338 total characters:
        All characters are of type 'unord'
        All characters have equal weight
        4709 characters are constant (proportion = 0.882166)
        14 variable characters are parsimony-uninformative
        Number of parsimony-informative characters = 615
    Gaps are treated as "missing"
  Starting tree(s) obtained via stepwise addition
    Addition sequence: random
    Number of replicates = 10
    Starting seed = generated automatically
    Number of trees held at each step = 1
  Branch-swapping algorithm: tree-bisection-reconnection (TBR) with reconnection limit = 8
    Steepest descent option not in effect

    No more than 2 trees of score (length) greater than or equal to 5 will be saved in each replicate
  Initial 'Maxtrees' setting = 100 (will be auto-increased by 100)

  Branches collapsed (creating polytomies) if maximum branch length is zero
  'MulTrees' option in effect
  No topological constraints in effect
  Trees are unrooted

  1000 partition-homogeneity test replicates completed
  Time used = 00:03:57 (CPU time = 00:03:57.4)

Results of partition-homogeneity test:

        Sum of       Number of
  tree lengths       replicates
  -----------------------------

           749            4
           750            6
           751            5
           752           11
           753*          17
           754           43
           755           54
           756           69
           757          101
           758          125
           759          136
           760          107
           761          121
           762           83
           763           61
           764           30
           765           15
           766            9
           767            2
           768            1

              * = sum of lengths for original partition

   P value = 1 - (957/1000) = 0.043000

Thanks for pointing out the issue with the MaxTrees, that was helpful. 

Regards,

Jeremiah 

[Miller, Mark]

Awesome. Well done.