Dear Ziheng and/or Bruce
I am trying to run a bpp to estimate a joint species delimitation and species tree estimation for 14 species (316 individuals) with 4 loci. I have been trying to fix the ctl files but I keep getting an error that says increase the NS in the source file. I have change the NS number, however I still have the same issue any thoughts of what I am doing wrong?
Best,
Melissa
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I try to set up a species delimitation analysis with BPP 3.0 using 4 loci. 190 individuals were grouped into 40 populations and the guide tree was estimated with starBEAST. Prior to analysing the data, I wanted to run BPP without data on the prior only. I tried 16 combinations of 4 theta and 4 tau priors (G(2,20), G(2,200), G(2,2000), G(2,20000) each). I expected, that small prior values for tau and high values for theta will result in less support even for deeper nodes in the guide tree. However, even the most exteme values resulted in nearly identical lumping behaviour. Is my expectation wrong? One node I want to test always gets very high support values. I also played with the finetuning parameters for delimitation algorithm1 (alpha=2, mean=0.1 up to 10) but the before mentioned node still always gets high support.
Is this behaviour normal when running BPP under the prior only and how can I ensure, that deeper nodes are appropriately tested for support? The guide tree has 40 tips and is more or less balanced.
Thanks for your time!
Best,
Jonas
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
I ran BPP3.2a with example files (A00) and results showed output files including MCMC.text, out.txt, as well as tree file. However, when I ran my data with A11 model, the output did not have any tree file. I also did not find any tree file as output when example files (A11.bpp.ctl, frogs.txt, and frogs.Imap.txt) were run. Should I get or not tree file in results while using A11 model?
Next, I have seen that many published papers with BPP analyses used both algorithm (0 & 1) to get consistent results. I think these are two alternative rjMCMC algorithms that were also explained in tutorial; (1) algorithm '0' means- speciesdelimitation = 1 0 2. * speciesdelimitation algorithm0 and finetune(e) (2) algorithm '1' means- speciesdelimitation = 1 1 2 1 * speciesdelimitation algorithm1 finetune (a m). Both of them were used together in the example of control file ('A11.bpp.ctl' in frog folder).
Should I use them together in a control file or need to use only one algorithm at a time?
or
Is there something else that need to specify algorithm 0 & 1 in control file?
Thanks in advance
Last edit: Anonymous 2016-03-25
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
if you don't see a tree from running the tutorial, there is no tree.
algorithms 0 and 1 are mentioned in the documentation. also search the discssion site as they were discused before.
ziheng
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
Anonymous
-
2017-08-18
Post awaiting moderation.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello,
I am implemetning BPP 3.2 with the A10 model with species deliitation= 1 0 2 (algortihm 0 and fine tune 2). I am specifying a guide tree made in Beast. BPP reads my species, but continues to give me an error (expecting ; in the species tree).
My species&tree= 7 A B C D E F G
3 3 3 3 3 3 3
((A), B), C), D), E), F), G));
Followed the example files but wondering if formatting is off.
Thanks,
Chase
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I do want to conform is this control file and heredity. txt is ok for BPP analysis I always receive the message "EOF when reading rates from heredity.txt' while running BPP. I am using one mtDNA and four autosomal nDNA and one X linked chromosome.
I'm doing a A10 analysis, but it is constantly stopping after the first step, i.e., it analyses the first possibility (say 1111100000) and it freezes on the second (1111111111)
Cheers
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have very basic question. One of the reviewer has asked me to run BPP for species tree delimitation and construction.
I have 100 loci of of 6 species, in which only 2 species have 3 individuals and rest of
the 4 species have one indiviudal. By reading, the manual I understood that it is not possible to run BPP in my case since I have only one individual for 4 species and it is not possible to calculate the population theta paramter.
Please let me know if it is possible to run the BPP in my case on 100 loci and the model A11.
thanks
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
it should be fine.
if a species never has 2 or more sequences at the same locus, it wont be pissible to estimate theta for that species, but you can still run bpp.
do the tutorial in yang 2015 CURRNT ZOOLOGY first.
ziheng
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Dear all,
I am running BP&P on my computer normally. I try to instal in on cluster, analysis is running, but I don't get results because of this error:
Error: check #seqs and tree: perhaps too many '('?.
Thank you in advance! :)
Best,
Tatiana
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I plan to use BPP v3 for species delimitation using >100 loci (could go up to 2500) derived from sequencing ultraconserved genomic elements (see Faircloth et al. 2012) in > 100 individuals. I understand that it's not possible to do it all in one run. One could of course analyze distinct portions of individuals (e.g. those belonging to a distinct clade) in separate runs, e.g. 25 individuals per run.
My questions:
- What is your experience regarding the upper limit of number of loci (each ~1000 bp long, with 2-10% nucleotide diversity), if say 25 Individuals are delimited simulatneously (and simultaneously with the guide tree)?
- Since I can work on a multi-core machine: can parallelization be performed and if yes, how is it done (e.g. across regions?)
Thank you for your help
Simon
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
its hard to be specific about the limits of the program.
you can prepare the files and if they dont run, split the data.
With the A01 analysis, we managed to run the program with 1000 and 10000 loci, each of 17 sequences.
best,
ziheng
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
Anonymous
-
2018-07-11
Post awaiting moderation.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
Anonymous
-
2018-07-11
Post awaiting moderation.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
Anonymous
-
2018-07-13
Post awaiting moderation.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
View and moderate all "BP&P Q&A" comments posted by this user
Mark all as spam, and block user from posting to "Discussion"
ok - thanks, that worked. Thanks for taking the time to answer!
View and moderate all "BP&P Q&A" comments posted by this user
Mark all as spam, and block user from posting to "Discussion"
Dear Ziheng and/or Bruce
I am trying to run a bpp to estimate a joint species delimitation and species tree estimation for 14 species (316 individuals) with 4 loci. I have been trying to fix the ctl files but I keep getting an error that says increase the NS in the source file. I have change the NS number, however I still have the same issue any thoughts of what I am doing wrong?
Best,
Melissa
Please email me your control file, sequence data file and map file (brannala@ucdavis.edu) and I will have a look.
View and moderate all "BP&P Q&A" comments posted by this user
Mark all as spam, and block user from posting to "Discussion"
Hi,
I try to set up a species delimitation analysis with BPP 3.0 using 4 loci. 190 individuals were grouped into 40 populations and the guide tree was estimated with starBEAST. Prior to analysing the data, I wanted to run BPP without data on the prior only. I tried 16 combinations of 4 theta and 4 tau priors (G(2,20), G(2,200), G(2,2000), G(2,20000) each). I expected, that small prior values for tau and high values for theta will result in less support even for deeper nodes in the guide tree. However, even the most exteme values resulted in nearly identical lumping behaviour. Is my expectation wrong? One node I want to test always gets very high support values. I also played with the finetuning parameters for delimitation algorithm1 (alpha=2, mean=0.1 up to 10) but the before mentioned node still always gets high support.
Is this behaviour normal when running BPP under the prior only and how can I ensure, that deeper nodes are appropriately tested for support? The guide tree has 40 tips and is more or less balanced.
Thanks for your time!
Best,
Jonas
View and moderate all "BP&P Q&A" comments posted by this user
Mark all as spam, and block user from posting to "Discussion"
Hi,
I ran BPP3.2a with example files (A00) and results showed output files including MCMC.text, out.txt, as well as tree file. However, when I ran my data with A11 model, the output did not have any tree file. I also did not find any tree file as output when example files (A11.bpp.ctl, frogs.txt, and frogs.Imap.txt) were run. Should I get or not tree file in results while using A11 model?
Next, I have seen that many published papers with BPP analyses used both algorithm (0 & 1) to get consistent results. I think these are two alternative rjMCMC algorithms that were also explained in tutorial; (1) algorithm '0' means- speciesdelimitation = 1 0 2. * speciesdelimitation algorithm0 and finetune(e) (2) algorithm '1' means- speciesdelimitation = 1 1 2 1 * speciesdelimitation algorithm1 finetune (a m). Both of them were used together in the example of control file ('A11.bpp.ctl' in frog folder).
Should I use them together in a control file or need to use only one algorithm at a time?
or
Is there something else that need to specify algorithm 0 & 1 in control file?
Thanks in advance
Last edit: Anonymous 2016-03-25
if you don't see a tree from running the tutorial, there is no tree.
algorithms 0 and 1 are mentioned in the documentation. also search the discssion site as they were discused before.
ziheng
View and moderate all "BP&P Q&A" comments posted by this user
Mark all as spam, and block user from posting to "Discussion"
Hello,
I am implemetning BPP 3.2 with the A10 model with species deliitation= 1 0 2 (algortihm 0 and fine tune 2). I am specifying a guide tree made in Beast. BPP reads my species, but continues to give me an error (expecting ; in the species tree).
My species&tree= 7 A B C D E F G
3 3 3 3 3 3 3
((A), B), C), D), E), F), G));
Followed the example files but wondering if formatting is off.
Thanks,
Chase
Your newick tree is incorrect. Try this:
species&tree= 7 A B C D E F G
3 3 3 3 3 3 3
((((((A, B), C), D), E), F), G);
View and moderate all "BP&P Q&A" comments posted by this user
Mark all as spam, and block user from posting to "Discussion"
I do want to conform is this control file and heredity. txt is ok for BPP analysis I always receive the message "EOF when reading rates from heredity.txt' while running BPP. I am using one mtDNA and four autosomal nDNA and one X linked chromosome.
first analyze the loci from the mt, nuclear and X separately.
EOF means end of file. you need as many numbers in the file as you have loci.
ziheng
View and moderate all "BP&P Q&A" comments posted by this user
Mark all as spam, and block user from posting to "Discussion"
Hi,
I'm doing a A10 analysis, but it is constantly stopping after the first step, i.e., it analyses the first possibility (say 1111100000) and it freezes on the second (1111111111)
Cheers
Hi Ziheng
I have very basic question. One of the reviewer has asked me to run BPP for species tree delimitation and construction.
I have 100 loci of of 6 species, in which only 2 species have 3 individuals and rest of
the 4 species have one indiviudal. By reading, the manual I understood that it is not possible to run BPP in my case since I have only one individual for 4 species and it is not possible to calculate the population theta paramter.
Please let me know if it is possible to run the BPP in my case on 100 loci and the model A11.
thanks
it should be fine.
if a species never has 2 or more sequences at the same locus, it wont be pissible to estimate theta for that species, but you can still run bpp.
do the tutorial in yang 2015 CURRNT ZOOLOGY first.
ziheng
View and moderate all "BP&P Q&A" comments posted by this user
Mark all as spam, and block user from posting to "Discussion"
Dear all,
I am running BP&P on my computer normally. I try to instal in on cluster, analysis is running, but I don't get results because of this error:
Error: check #seqs and tree: perhaps too many '('?.
Thank you in advance! :)
Best,
Tatiana
In this case the error mesasge is clear enough. something is wrong with the tree you supplied in the control file.
ziheng
View and moderate all "BP&P Q&A" comments posted by this user
Mark all as spam, and block user from posting to "Discussion"
Hi Ziheng / Bruce
I plan to use BPP v3 for species delimitation using >100 loci (could go up to 2500) derived from sequencing ultraconserved genomic elements (see Faircloth et al. 2012) in > 100 individuals. I understand that it's not possible to do it all in one run. One could of course analyze distinct portions of individuals (e.g. those belonging to a distinct clade) in separate runs, e.g. 25 individuals per run.
My questions:
- What is your experience regarding the upper limit of number of loci (each ~1000 bp long, with 2-10% nucleotide diversity), if say 25 Individuals are delimited simulatneously (and simultaneously with the guide tree)?
- Since I can work on a multi-core machine: can parallelization be performed and if yes, how is it done (e.g. across regions?)
Thank you for your help
Simon
its hard to be specific about the limits of the program.
you can prepare the files and if they dont run, split the data.
With the A01 analysis, we managed to run the program with 1000 and 10000 loci, each of 17 sequences.
best,
ziheng