failed in modelfinder

[YanYan]

It works in 3 gene matrix，but failed in 62 genes. Is there anyone who have time to give me some suggestion? Thanks!

Host:    DESKTOP-4QC0DVO (AVX2, FMA3, 63 GB RAM)
Command: bin\iqtree -s n20dal.phy -m MF -nt AUTO
Seed:    134096 (Using SPRNG - Scalable Parallel Random Number Generator)
Time:    Fri Jun 28 16:03:17 2019
Kernel:  AVX+FMA - auto-detect threads (12 CPU cores detected)

Reading alignment file n20dal.phy ... Phylip format detected
Alignment most likely contains protein sequences
WARNING: 1 sites contain only gaps or ambiguous characters.
Alignment has 20 sequences with 47119 columns, 14598 distinct patterns
25073 parsimony-informative, 7627 singleton sites, 14419 constant sites
            Gap/Ambiguity  Composition  p-value
   1  FauNEOPT     13.33%    failed      0.00%
   2  LfoCHILO     13.10%    failed      0.00%
   3  LpoXIPHOS    13.13%    failed      0.00%
   4  MgaARACH     13.21%    failed      0.00%
   5  NamDIPLO     13.51%    failed      0.00%
   6  OstOST       13.01%    failed      0.00%
   7  PaqCOLL      13.37%    failed      0.11%
   8  StuREMI      13.20%    failed      0.00%
   9  DmaBRANCH    13.10%    passed     84.89%
  10  HmaCEPHAL    12.32%    failed      0.00%
  11  ScolCHILO    13.37%    failed      0.00%
  12  SpoCHILO     13.44%    failed      0.00%
  13  AmaDIPLO     13.26%    passed     99.24%
  14  Han2SYMPH    13.42%    failed      0.00%
  15  Pge2DIPLO    12.82%    failed      0.00%
  16  Pol2DIPLO    13.04%    failed      0.00%
  17  Scu3SYMPH    13.21%    failed      0.00%
  18  Col2PYCNO    12.93%    failed      0.00%
  19  Ctas2CHILO   13.32%    failed      0.00%
  20  EuryPAURO    13.32%    failed      0.00%
****  TOTAL        13.17%  18 sequences failed composition chi2 test (p-value<5%; df=19)

Create initial parsimony tree by phylogenetic likelihood library (PLL)... 0.109 seconds
NOTE: ModelFinder requires 451 MB RAM!
ModelFinder will test 546 protein models (sample size: 47119) ...
 No. Model         -LnL         df  AIC          AICc         BIC
Measuring multi-threading efficiency up to 12 CPU cores
Increase to 10 rounds for branch lengths
21 trees examined
Threads: 1 / Time: 12.341 sec / Speedup: 1.000 / Efficiency: 100% / LogL: -935133
(FauNEOPT:0.0861668353,(((((LfoCHILO:0.0526173546,(ScolCHILO:0.0611790508,(SpoCHILO:0.0761707159,Ctas2CHILO:0.1122432335):0.0165150318):0.0106031376):0.0127779567,(((LpoXIPHOS:0.0933324633,MgaARACH:0.1038009172):0.0454104437,(OstOST:0.1393593461,StuREMI:0.1163944343):0.0376380640):0.0116898091,NamDIPLO:0.0553629371):0.0242750003):0.0198689141,(((AmaDIPLO:0.0676351086,(Pge2DIPLO:0.1242395492,Pol2DIPLO:0.1532807224):0.0302866278):0.0272275184,(Han2SYMPH:0.0735391199,Scu3SYMPH:0.0729860441):0.0279327799):0.0367403485,EuryPAURO:0.1491317583):0.0412090034):0.0380166039,(PaqCOLL:0.1560448170,Col2PYCNO:0.1530822207):0.0585503395):0.0247971878,DmaBRANCH:0.1201145706):0.0645290081,HmaCEPHAL:0.3752526841);
new_tree_lh: -865423.1692921221   tree_lh: -863818.5083998179
ERROR: C:\Users\minh\Dropbox\iqtree-git\tree\phylotree.cpp:2652: virtual double PhyloTree::optimizeAllBranches(int, double, int): Assertion `fabs(new_tree_lh-tree_lh) < max_delta_lh' failed.
ERROR:
ERROR: *** IQ-TREE CRASHES WITH SIGNAL ABORTED
ERROR: *** For bug report please send to developers:
ERROR: ***    Log file: n20dal.phy.log
ERROR: ***    Alignment files (if possible)

[Minh Bui]

Hi YanYan,

I could not replicate this crash using the latest version 1.6.11. 

Which version did you use? Can you pls try the latest version.

Cheers,

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.
To view this discussion on the web visit https://groups.google.com/d/msgid/iqtree/c868e722-2307-4e8a-9730-3a5f7723892d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
<n20dal.phy>

[YanYan]

Dear Minh:

Thank you very much to give me suggestion so soon.

Yes I using latest version 1.6.11, and test in win10 and win7.

The n20dal.phy file I tried to analysis is a degenerated matrix, that is a coding method for nucleotides in which codons in a data matrix are completely degenerated, for the first and third codon positions using ambiguity coding e.g., CAT -- . CAY and TTA--. YTN.

But IQ-tree detected: Alignment most likely contains protein sequences (not DNA), and try to use protein model to analysis and then failed.

I do not know whether it is the reason that –m MF requirement failed.

I test an amino acid matrix. It works after I enter iqtree –s aamatrix.phy

Is it possible to let IQ-tree recognize degenerated DNA matrix?

在 2019年6月29日星期六 UTC+8上午11:20:41，YanYan写道：

[Minh Bui]

Ah I see. That’s the reason why IQ-TREE mis-identify your sequences as protein. You can use -st DNA to enforce DNA sequences, instead.

Cheers,

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.

To view this discussion on the web visit https://groups.google.com/d/msgid/iqtree/c3064706-578e-4b71-ba6f-36f40c47ab63%40googlegroups.com.

[YanYan]

Dear Minh:

After adding -st DNA, IQ-tree works! Thank you! I am very happy that it is really fast.

Cheers