PAMLX 4.7 MCMCTree questions

[Anna Kusnezowa]

Dear all,

I try to estimate divergence times of proteins from the Thioredoxin family.

For do this I would use a routine from 

A step-by-step manual for MCMCtree with an example dataset (MCMCtree.example.tgz) .

I have problems with estimation of the gradient and Hessian

Comment: although the Manual recommend to use unrooted tree without the clock for gradient and Hesssian estimation it is not possible to use MCMCTree- module with unrooted tree .  

with unrooted tree 

/Users/leichertlab/bioinfo-programms/Contents/paml4.7/bin/mcmctree   running...

Error: check and think about multificating trees..

Unfortunately MCMCtree can not read example MSA-file although the CodeML has read without any errors and I have got the expected substitution rate is per time unit

    0.047131 (Manual).

/Users/leichertlab/bioinfo-programms/Contents/paml4.7/bin/mcmctree   running...

MCMCTREE in paml version 4.7, January 2013

Reading options from mcmctree.ctl.tmp..

Reading master tree.

(((Cyprinus_carpio, (Polymixia_japonica, Tetraodon_nigroviridis)), (Protopterus_aethiopicus, (Xenopus_tropicalis, (Struthio_camelus, (Ornithorhynchus_anatinus, Homo_sapiens))))), (((Plesiobatis_daviesi, (Amblyraja_radiata, Okamejei_kenojei)), (Squalus_acanthias, (Heterodontus_francisci, (Scyliorhinus_canicula, Mustelus_manazo)))), ((Callorhinchus_callorynchus, (Callorhinchus_milii, Callorhinchus_capensis)), ((Hariotta_raleighana, Rhinochimaera_pacifica), (Hydrolagus_lemures, (Chimaera_sp, Chimaera_monstrosa))))));

Reading sequence data..  2 loci

*** Locus 1 ***

ns = 23   ls = 3248

Reading sequences, sequential format..

Reading seq # 1: Cyprinus_carpio     

Error in sequence data file: L at 2 seq 1.

Make sure to separate the sequence from its name by 2 or more spaces.

/Users/leichertlab/bioinfo-programms/Contents/paml4.7/bin/mcmctree   finished

My research has shown that, MCMCTree can work only with interleaved format of MSA.

Is there a way to read a sequential files?

And my second question

I have try to estimate divergence time for my data. I have 21 AA-sequences in MSA. CodeML has work without any errors. MCMCTree has read my interleaved MSA file and has processsed it. But I have got zero values in Blocks

Users/leichertlab/bioinfo-programms/Contents/paml4.7/bin/mcmctree running...

MCMCTREE in paml version 4.7, January 2013

Reading options from mcmctree.ctl.tmp..

Reading master tree.

(((((((((142035511/1-128, ((142745684/1-129, 134344726/1-147), 142438332/1-136)), 136582819/1-133), 135289905/1-170), (/1-115, (139111483/1-115, (140401181/1-119, 136236388/1-147)))), (143943928/1-128, 143027041/1-206)), (139525871/1-116, 135374423/1-217)), 134901086/1-156), 136096735/1-162), (135544395/1-108, (137293039/1-109, (EG11031/1-109, (142712172/1-138, 134462067/1-160)))));

unrecognized fossil calibration: @4.2

Reading sequence data.. 2 loci

*** Locus 1 ***

ns = 21 ls = 143

Reading sequences, interlaved format..

Reading seq # 1: 142035511/1-128

Reading seq # 2: 142745684/1-129

Reading seq # 3: 134344726/1-147

Reading seq # 4: 142438332/1-136

Reading seq # 5: 143943928/1-128

Reading seq # 6: 143027041/1-206

Reading seq # 7: 139525871/1-116

Reading seq # 8: 135374423/1-217

Reading seq # 9: 137293039/1-109

Reading seq #10: 135544395/1-108

Reading seq #11: EG11031/1-109

Reading seq #12: 142712172/1-138

Reading seq #13: 134462067/1-160

Reading seq #14: 142888794/1-115

Reading seq #15: 135289905/1-170

Reading seq #16: 136096735/1-162

Reading seq #17: 139111483/1-115

Reading seq #18: 136582819/1-133

Reading seq #19: 140401181/1-119

Reading seq #20: 134901086/1-156

Reading seq #21: 136236388/1-147

block 1: 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

/Users/leichertlab/bioinfo-programms/Contents/paml4.7/bin/mcmctree finished

And I have no output information in out-file

Is it necessary to enter any calibration times into the tree?

Can I estimate divergence without any calibration times? I have found no calibration times for thioredoxine.

I thank you in advance,

Anna 

[Ziheng]

there is a tutorial about mcmctree at the paml web site.
the program can read both interleaved and sequential formats. your problem has to do with the sequence alignment file and tree file. One possibility is that the problem is due to your sequences starting with numbers. You can add a letter at the beginning of each name and see whether that fixes the problem.

Reading seq # 1: Cyprinus_carpio
Error in sequence data file: L at 2 seq 1.
Make sure to separate the sequence from its name by 2 or more spaces.

Also the doc is in the file pamlDOC.pdf.

you can't estimate divergence times without calibrations.

ziheng

[Ales Bucek]

I had identical "Make sure to separate the sequence from its name by 2 or more spaces." error.

In my case, the species name was separated by tab from the sequence, replacing the tab by two spaces (e.g. in bash via sed 's/\t/  /') fixed the problem.

Best,

Ales