Version 1.4.3-heterotachy for Mac OS: IQ-TREE CRASHES WITH SIGNAL ABORTED

[Ajith Harish]

Hello Minh,

IQ-TREE heterotachy version crashed with my data set. But the same data worked just fine with IQ-TREE version 1.5.4 for Mac OS X 64-bit. I use OS X 10.10.5 and my alignment is binary data (presence/absence matrix). 

I got the same error with both the multi-core version and the serial version. As you will see this happens after optimizations, which took over 4 hours (multi-core version). The relevant parts of the log is as follows,

IQ-TREE multicore version 1.4.3-heterotachy for Mac OS X 64-bit built Aug 12 2016

Copyright (c) 2011-2015 Nguyen Lam Tung, Olga Chernomor, Arndt von Haeseler and Bui Quang Minh.

Host:    myroom.local (AVX, 8 GB RAM)

Command: iqtree-omp-ht -s Eq_47_Occ.IQ.phy -m GTR2+FO+H4 -nt 4

Seed:    297054 (Using SPRNG - Scalable Parallel Random Number Generator)

Time:    Sun Apr 23 22:54:30 2017

Kernel:  AVX - 4 threads (4 CPU cores detected)

..................

.................

................

Heterotachy mixture model name: MIX{GTR2,GTR2,GTR2,GTR2}+FO

Initializing heterotachy model with 4 mixture branch lengths

Creating fast SIMD initial parsimony tree by random order stepwise addition...

0.028 seconds, parsimony score: 13911 (based on 1542 informative sites)

NOTE: 13 MB RAM is required!

Estimate model parameters (epsilon = 0.100)

tree_lh = -5388.652

1. Initial log-likelihood: -16584.246

Optimal log-likelihood: -31875.441

Mixture weights: 0.543 0.351 -0.225 0.331

Parameters optimization took 1 rounds (16966.252 sec)

--------------------------------------------------------------------

|             INITIALIZING CANDIDATE TREE SET                      |

--------------------------------------------------------------------

Computing ML distances based on estimated model parameters... 0.467 sec

Computing BIONJ tree...

0.299 seconds

Log-likelihood of BIONJ tree: -24424.529

Generating 98 parsimony trees... 1.860 seconds 100 distinct starting trees

Computing log-likelihood of 100 initial trees ... 25.982 seconds

Optimizing top 20 initial trees with NNI...

STACK TRACE FOR DEBUGGING:

1   std::vector<std::vector<int, std::allocator<int> >, std::allocator<std::vector<int, std::allocator<int> > > >::reserve(unsigned long)

2   _sigtramp()

6   double PhyloTree::computeSitemodelLikelihoodFromBufferEigenSIMD<Vec2d, 2, 20>()

7   double** aligned_alloc<double*>(unsigned long)

8   BioNj::create(char const*, char const*)

9   std::vector<NNIMove, std::allocator<NNIMove> >::clear()

10   CandidateSet::operator=(CandidateSet const&)

11   bool Checkpoint::get<std::string>(std::string, std::string&)

12   bool Checkpoint::get<std::string>(std::string, std::string&)

13   std::vector<ModelInfo, std::allocator<ModelInfo> >::~vector()

14   std::vector<PartitionInfo, std::allocator<PartitionInfo> >::operator=(std::vector<PartitionInfo, std::allocator<PartitionInfo> > const&)

15   std::vector<std::vector<int, std::allocator<int> >, std::allocator<std::vector<int, std::allocator<int> > > >::reserve(unsigned long)

*** IQ-TREE CRASHES WITH SIGNAL ABORTED

Could you please suggest what might be wrong here? Thanks.

Cheers,

Ajith

[Bui Quang Minh]

Dear Ajith,

Thanks for using this model. We haven’t tried the heterotachy model (+H) on binary data. Thus, can you pls send us the data for debugging purpose (via personal email!) ?

Minh

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

--

Bui Quang Minh

Center for Integrative Bioinformatics Vienna (CIBIV)

Campus Vienna Biocenter 5, VBC5, Ebene 1

A-1030 Vienna, Austria

Phone: ++43 1 4277 74326

Email: minh.bui (AT) univie.ac.at