HyPhy message board
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl
Methodology Questions >> How to >> yangnielsienbranchsite2005
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?num=1354636868

Message started by zzs on Dec 4th, 2012 at 8:01am

Title: yangnielsienbranchsite2005
Post by zzs on Dec 4th, 2012 at 8:01am
I have a dataset with two virus lineages, one lineage (A) has 130 sequences (650 codon sites) while the other lineage (B) has only one sequences. no recombination detected by SBP. 1) will it be possible using this dataset to estimate episodic selection on lineage B?
2) what method should be better? I tried branchsite REL in HYpHy but run into error and quit at phase 1. But i successfully ran yangnielsienbranchsite2005 method using alternative model. In the output of yangnielsienbranchsite2005, there are four classes of probability values for each site, I tried to use the 'most likely category assignment' option to decide whether some site is under episodic selection, will this be right? How HyPHy decide which category each site belongs to ? For the yang2005 method, the P value of the four classes (1-4) for each site were given, but the assignment method assign (0, 1 ,2 or 3 class) to each site, dose this 0-3 correspond to the 1-4 classes of the yang method? In the paper of Yang 2005, there are two bayes methods used: na ̈ıve empirical Bayes and Bayes empirical Bayes. do these methods included in HyPhy? 
Thanks.

Title: Re: yangnielsienbranchsite2005
Post by konrad on Dec 4th, 2012 at 8:10am
Dear zzs,

We recommend branch-site REL for this; given that you are only interested in one lineage, you can use the p-values directly without needing to correct for multiple testing. Please provide more information about the error you encountered using branch-site REL.

We do _not_ recommend the branch-site test of Yang and Nielsen, for reasons explained in our branch-site REL paper: specifically, the assumption of homogeneity among the background branches can lead to incorrect inference, as demonstrated by our simulation study in that paper.

Hope this helps,
Konrad Scheffler

Title: Re: yangnielsienbranchsite2005
Post by zzs on Dec 4th, 2012 at 9:04am
Thanks for your help, the errors will be posted soon. Another question, if using yang 2005 method, the P-values for the four classes are similar some time or p<0.001 for all the four classes, how should i decide which class the site belongs to?

Title: mac crash report when running branchsite REL in hyphy
Post by zzs on Dec 4th, 2012 at 5:48pm
Process:         HyPhy [35903]
Path:            /Applications/HyPhy.app/Contents/MacOS/HyPhy
Identifier:      org.hyphy
Version:         2.10beta (2.10beta)
Code Type:       X86 (Native)
Parent Process:  launchd [234]

Date/Time:       2012-12-04 19:42:44.383 -0600
OS Version:      Mac OS X 10.7.5 (11G63)
Report Version:  9

Interval Since Last Report:          2266281 sec
Crashes Since Last Report:           32
Per-App Interval Since Last Report:  372346 sec
Per-App Crashes Since Last Report:   10
Anonymous UUID:                      6C22E135-33D2-4E82-8474-8B2682CF2767

Crashed Thread:  0  Dispatch queue: com.apple.main-thread

Exception Type:  EXC_CRASH (SIGABRT)
Exception Codes: 0x0000000000000000, 0x0000000000000000

Application Specific Information:
objc[35903]: garbage collection is OFF

Thread 0 Crashed:: Dispatch queue: com.apple.main-thread
0   libsystem_kernel.dylib             0x9355183e __psynch_cvwait + 10
1   libsystem_c.dylib                  0x9b5b4e78 _pthread_cond_wait + 914
2   libsystem_c.dylib                  0x9b55c82a pthread_cond_wait + 48
3   org.hyphy                           0x000060ac gomp_sem_wait + 60
4   org.hyphy                           0x00006195 gomp_barrier_wait_end + 101
5   org.hyphy                           0x000061f7 gomp_barrier_wait + 55
6   org.hyphy                           0x00005e7f gomp_team_end + 47

Thread 1:: Dispatch queue: com.apple.libdispatch-manager
0   libsystem_kernel.dylib             0x9355290a kevent + 10
1   libdispatch.dylib                  0x9c384e04 _dispatch_mgr_invoke + 969
2   libdispatch.dylib                  0x9c383853 _dispatch_mgr_thread + 53

Thread 2:: com.apple.CFSocket.private
0   libsystem_kernel.dylib             0x93551b42 __select + 10
1   com.apple.CoreFoundation            0x9b712e15 __CFSocketManager + 1557
2   libsystem_c.dylib                  0x9b5b0ed9 _pthread_start + 335
3   libsystem_c.dylib                  0x9b5b46de thread_start + 34

Thread 3:
0   libsystem_kernel.dylib             0x9355183e __psynch_cvwait + 10
1   libsystem_c.dylib                  0x9b5b4e78 _pthread_cond_wait + 914
2   libsystem_c.dylib                  0x9b55c82a pthread_cond_wait + 48
3   org.hyphy                           0x000060ac gomp_sem_wait + 60
4   org.hyphy                           0x00006195 gomp_barrier_wait_end + 101
5   org.hyphy                           0x000061f7 gomp_barrier_wait + 55
6   org.hyphy                           0x00005e34 gomp_thread_start + 260
7   libsystem_c.dylib                  0x9b5b0ed9 _pthread_start + 335
8   libsystem_c.dylib                  0x9b5b46de thread_start + 34

Thread 4:
0   libsystem_kernel.dylib             0x9355183e __psynch_cvwait + 10
1   libsystem_c.dylib                  0x9b5b4e78 _pthread_cond_wait + 914
2   libsystem_c.dylib                  0x9b55c82a pthread_cond_wait + 48
3   org.hyphy                           0x000060ac gomp_sem_wait + 60
4   org.hyphy                           0x00006195 gomp_barrier_wait_end + 101
5   org.hyphy                           0x000061f7 gomp_barrier_wait + 55
6   org.hyphy                           0x00005e3c gomp_thread_start + 268
7   libsystem_c.dylib                  0x9b5b0ed9 _pthread_start + 335
8   libsystem_c.dylib                  0x9b5b46de thread_start + 34

Thread 5:
0   libsystem_kernel.dylib             0x9355183e __psynch_cvwait + 10
1   libsystem_c.dylib                  0x9b5b4e78 _pthread_cond_wait + 914
2   libsystem_c.dylib                  0x9b55c82a pthread_cond_wait + 48
3   org.hyphy                           0x000060ac gomp_sem_wait + 60
4   org.hyphy                           0x00006195 gomp_barrier_wait_end + 101
5   org.hyphy                           0x000061f7 gomp_barrier_wait + 55
6   org.hyphy                           0x00005e3c gomp_thread_start + 268
7   libsystem_c.dylib                  0x9b5b0ed9 _pthread_start + 335
8   libsystem_c.dylib                  0x9b5b46de thread_start + 34

Thread 6:
0   libsystem_kernel.dylib             0x9355202e __workq_kernreturn + 10
1   libsystem_c.dylib                  0x9b5b2ccf _pthread_wqthread + 773
2   libsystem_c.dylib                  0x9b5b46fe start_wqthread + 30

Thread 0 crashed with X86 Thread State (32-bit):
  eax: 0x00000000  ebx: 0x00000000  ecx: 0x

Title: crash complete version
Post by zzs on Dec 4th, 2012 at 5:50pm
let me know if further information is needed.
The program crash at branch estimation phase.
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?action=downloadfile;file=hyphyerror.txt (45 KB | 0 )

Title: Re: mac crash report when running branchsite REL in hyphy
Post by Sergei on Dec 5th, 2012 at 2:35am
Hi zzs,

This is an (unfortunate) known issue with OS 10.7 and multithreading (it affects not just HyPhy but many other programs that use OpenMP). See Multimedia File Viewing and Clickable Links are available for Registered Members only!!  You need to Login Login for a fix.

Sergei

HyPhy message board » Powered by YaBB 2.5.2!
YaBB Forum Software © 2000-2024. All Rights Reserved.