ERROR: File not found GTR

[xiaomin liu]

Hello all,

I am running iqtree v1.6.10 to calculate a snp dataset, but it keeps reporting errors.

command ：
$ /ds3200_1/proc/iqtree-1.6.10-Linux/bin/iqtree -s all.LDfilter.thinned.sort.sequences.phy -nt 10 -m GTR+ASC -bb 1000 -pre all.LDfilter. thinned -bnni -redo


This dataset was obtained  from the original dataset via the thin command of vcftools.

Error log:

IQ-TREE multicore version 1.6.10 for Linux 64-bit built Feb 19 2019
Developed by Bui Quang Minh, Nguyen Lam Tung, Olga Chernomor,
Heiko Schmidt, Dominik Schrempf, Michael Woodhams.

Host:    node01 (AVX2, FMA3, 6047 GB RAM)
Command: /ds3200_1/proc/iqtree-1.6.10-Linux/bin/iqtree -s all.LDfilter.thinned.sort.sequences.phy -nt 10 -m GTR+ASC -bb 1000 -pre all.LDfilter.thinned -bnni -redo
Seed:    495099 (Using SPRNG - Scalable Parallel Random Number Generator)
Time:    Sat Jul  9 13:54:11 2022
Kernel:  AVX+FMA - 10 threads (88 CPU cores detected)

Reading alignment file all.LDfilter.thinned.sort.sequences.phy ... Phylip format detected
Alignment most likely contains protein sequences
Alignment has 470 sequences with 160684 columns, 157215 distinct patterns
160684 parsimony-informative, 0 singleton sites, 0 constant sites
            Gap/Ambiguity  Composition  p-value
   1  10110         0.00%    failed      0.00%
   2  2020001       0.00%    failed      0.00%

...........

..........

****  TOTAL         0.00%  470 sequences failed composition chi2 test (p-value<5%; df=10)
NOTE: minimal branch length is reduced to 0.000000622339 for long alignment
NOTE: State(s) D, Q, E, H, I, L, F, P, V not present in alignment and thus removed from Markov process to prevent numerical problems
NOTE: 9 states (see above) are not present and thus removed from Markov process to prevent numerical problems

Create initial parsimony tree by phylogenetic likelihood library (PLL)... 569.444 seconds
Generating 1000 samples for ultrafast bootstrap (seed: 495099)...
ERROR: File not found GTR

Also, I tried iqtree2 running the full dataset (above 3 million) and the same error was reported.

log:

IQ-TREE multicore version 2.1.2 COVID-edition for Linux 64-bit built Mar 30 2021
Developed by Bui Quang Minh, James Barbetti, Nguyen Lam Tung,
Olga Chernomor, Heiko Schmidt, Dominik Schrempf, Michael Woodhams.

Host:    node01 (AVX2, FMA3, 6047 GB RAM)
Command: iqtree2 -s sequences.phy -T 10 -m GTR+ASC --alrt 1000 -B 1000 -mem 1600G -pre out4 -bnni -redo
Seed:    444924 (Using SPRNG - Scalable Parallel Random Number Generator)
Time:    Thu Jul  7 23:25:42 2022
Kernel:  AVX+FMA - 10 threads (88 CPU cores detected)

Reading alignment file sequences.phy ... Phylip format detected
Alignment most likely contains protein sequences
Alignment has 470 sequences with 3309536 columns, 3196027 distinct patterns
3309349 parsimony-informative, 187 singleton sites, 0 constant sites
            Gap/Ambiguity  Composition  p-value
   1  10110         0.00%    failed      0.00%
   2  2020001       0.00%    failed      0.00%

..........

..........

 470  ldz2020037    0.00%    failed      0.00%
****  TOTAL         0.00%  470 sequences failed composition chi2 test (p-value<5%; df=10)
NOTE: minimal branch length is reduced to 0.000000030216 for long alignment
NOTE: State(s) D, Q, E, H, I, L, F, P, V not present in alignment and thus removed from Markov process to prevent numerical problems
NOTE: 9 states (see above) are not present and thus removed from Markov process to prevent numerical problems

Create initial parsimony tree by phylogenetic likelihood library (PLL)... 22022.449 seconds
Generating 1000 samples for ultrafast bootstrap (seed: 444924)...
ERROR: File not found GTR

So,can you tell me  how to solve this problem? 

thanks all

[Lin Lin]

I also encountered a similar problem, I think it may be because of "Alignment most likely contains protein sequences"? What should I do to get iqtree to recognize my data as DNA?

[Thomas Wong]

Hi,

You can use the option "-st DNA" to specify the data type as DNA.

--
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 view this discussion on the web visit https://groups.google.com/d/msgid/iqtree/43422f89-a544-43ac-b191-c448f9d56fd3n%40googlegroups.com.

[Bui Quang Minh]

Hi Lin, The issue is that IQ-TREE tries to guess the type of sequences
by counting the fraction of A, C, G and T in your alignment. Only if
this fraction is larger than a threshold (I believe something like
90%) then it's DNA. Otherwise, it may miss-classify your sequence
type.

If you can use the option Thomas pointed out (-st DNA) to enforce it to be DNA.

Cheers
Minh

On Mon, Apr 8, 2024 at 5:19 PM Lin Lin <linli...@gmail.com> wrote:
>