sanderdebacker
commented
1 year ago Your dataset contains SNP data, which is DNA.
IQ-Tree states "Alignment most likely contains protein sequences".
The autodetect doesn't recognize your input as DNA and therefore gives an error about the GTR model.
Try adding the -st DNA argument to your command.
Hope this helps!
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