BSP analysis: Fatal exception during plot:null

500 views
Skip to first unread message

Jason

unread,
Dec 30, 2019, 1:56:55 PM12/30/19
to beast-users
Dear Beast-uers,

I am trying to use Tracer to perform the bayesian skyline reconstruction analysis, however, I always get the message " Fatal exception during plot:null".
The .log file  and .tree files are combined from several independent runnings with Beast v1.10.4. I have tried to use different version of tracer (Tracer V1.5, V1.6 and V1.7) to do the analysis, however none of them is working, did anyone have the same problem or good suggestions? Thanks in advance for all your kind help !



Best regards,

Aly Wala

unread,
Jan 6, 2020, 4:59:20 PM1/6/20
to beast-users
I also am having the same problem.

HS

unread,
Jan 7, 2020, 3:28:12 AM1/7/20
to beast-users
Hi,

You may check if the .trees file ends with "End;". It has to be in the end of the file.

Best,
Hovhannes

Alefiya Neemuchwala

unread,
Jan 7, 2020, 4:10:22 PM1/7/20
to beast...@googlegroups.com

Hi

I confirmed that my .trees file does end with “End;”. However my file is 4GB. Is this perhaps too big for my Windows computer.

Aly

--
You received this message because you are subscribed to the Google Groups "beast-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to beast-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/beast-users/b478fad7-05ec-4af6-81c8-328648b8035a%40googlegroups.com.

HS

unread,
Jan 8, 2020, 5:34:53 AM1/8/20
to beast-users
Hi Aly,

I don't know about the sizes. Intuitively, if your PC's RAM is 4GB, then loading 4GB .trees file could be problematic.

To figure it out you may use LogCombiner to re-sample the .log and .trees file less frequently and perform BS reconstruction with the thinner files.

Best,
Hovhannes 


On Tuesday, 7 January 2020 23:10:22 UTC+2, Aly Wala wrote:

Hi

I confirmed that my .trees file does end with “End;”. However my file is 4GB. Is this perhaps too big for my Windows computer.

Aly

 

From: beast...@googlegroups.com [mailto:beast...@googlegroups.com] On Behalf Of HS
Sent: January-07-20 3:28 AM
To: beast-users <beast...@googlegroups.com>
Subject: Re: BSP analysis: Fatal exception during plot:null

 

Hi,

 

You may check if the .trees file ends with "End;". It has to be in the end of the file.

 

Best,

Hovhannes

On Monday, 6 January 2020 23:59:20 UTC+2, Aly Wala wrote:

I also am having the same problem.

On Monday, 30 December 2019 13:56:55 UTC-5, Jason wrote:

Dear Beast-uers,

 

I am trying to use Tracer to perform the bayesian skyline reconstruction analysis, however, I always get the message " Fatal exception during plot:null".

The .log file  and .tree files are combined from several independent runnings with Beast v1.10.4. I have tried to use different version of tracer (Tracer V1.5, V1.6 and V1.7) to do the analysis, however none of them is working, did anyone have the same problem or good suggestions? Thanks in advance for all your kind help !

 

 

 

Best regards,

--
You received this message because you are subscribed to the Google Groups "beast-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to beast...@googlegroups.com.

Jiexiong Xie

unread,
Jan 8, 2020, 12:06:02 PM1/8/20
to beast...@googlegroups.com
Hi, Hovhannes,

Thanks for your suggestions, I also checked my files, they are indeed
end with "End" , my file is also quite heavy ( more than 10G) , I am
trying to use logcombiner to thiner the files, meanwhile,I am also
trying with the thiner files to do BS reconstruction, there are no
errors at the moment but it stuck there, there are no progress , maybe
I should wait much longer, or maybe I should just directly running my
file longer rather than combining different runnings.

Best regards,
Jason

2020-01-08 11:34 GMT+01:00, HS <hovo...@gmail.com>:
> Hi Aly,
>
> I don't know about the sizes. Intuitively, if your PC's RAM is 4GB, then
> loading 4GB .trees file could be problematic.
>
> To figure it out you may use LogCombiner to re-sample the .log and .trees
> file less frequently and perform BS reconstruction with the thinner files.
>
> Best,
> Hovhannes
>
> On Tuesday, 7 January 2020 23:10:22 UTC+2, Aly Wala wrote:
>>
>> Hi
>>
>> I confirmed that my .trees file does end with “End;”. However my file is
>> 4GB. Is this perhaps too big for my Windows computer.
>>
>> Aly
>>
>>
>>
>> *From:* beast...@googlegroups.com <javascript:> [mailto:
>> beast...@googlegroups.com <javascript:>] *On Behalf Of *HS
>> *Sent:* January-07-20 3:28 AM
>> *To:* beast-users <beast...@googlegroups.com <javascript:>>
>> *Subject:* Re: BSP analysis: Fatal exception during plot:null
>>
>>
>>
>> Hi,
>>
>>
>>
>> You may check if the .trees file ends with "End;". It has to be in the end
>>
>> of the file.
>>
>>
>>
>> Best,
>>
>> Hovhannes
>>
>> On Monday, 6 January 2020 23:59:20 UTC+2, Aly Wala wrote:
>>
>> I also am having the same problem.
>>
>> On Monday, 30 December 2019 13:56:55 UTC-5, Jason wrote:
>>
>> Dear Beast-uers,
>>
>>
>>
>> I am trying to use Tracer to perform the bayesian skyline reconstruction
>> analysis, however, I always get the message " Fatal exception during
>> plot:null".
>>
>> The .log file and .tree files are combined from several independent
>> runnings with Beast v1.10.4. I have tried to use different version of
>> tracer (Tracer V1.5, V1.6 and V1.7) to do the analysis, however none of
>> them is working, did anyone have the same problem or good suggestions?
>> Thanks in advance for all your kind help !
>>
>>
>>
>>
>>
>>
>>
>> Best regards,
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>>
>> "beast-users" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>>
>> email to beast...@googlegroups.com <javascript:>.
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/beast-users/b478fad7-05ec-4af6-81c8-328648b8035a%40googlegroups.com
>>
>> <https://can01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgroups.google.com%2Fd%2Fmsgid%2Fbeast-users%2Fb478fad7-05ec-4af6-81c8-328648b8035a%2540googlegroups.com%3Futm_medium%3Demail%26utm_source%3Dfooter&data=02%7C01%7Calefiya.neemuchwala%40oahpp.ca%7C725c42c41c724332fc9008d7934b8b1e%7Ccddc1229ac2a4b97b78a0e5cacb5865c%7C0%7C0%7C637139824975416478&sdata=Rt3F8aiwetQ5mp2ZABRofebCOJYtamWpJj3IIz7rS9Y%3D&reserved=0>
>> .
>>
>
> --
> You received this message because you are subscribed to the Google Groups
> "beast-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to beast-users...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/beast-users/1a60b24a-aa43-43a0-915c-a7953c596553%40googlegroups.com.
>

HS

unread,
Jan 9, 2020, 5:36:44 AM1/9/20
to beast-users
Hi Jason,

I had such experience with Tracer v1.7.1. If you use v1.7, you may try BS reconstruction with earlier versions. In my case v1.6 and v1.5 worked fine.

Best,
Hovhannes

Jiexiong Xie

unread,
Jan 9, 2020, 12:20:35 PM1/9/20
to beast...@googlegroups.com
Dear Hovhannes,

I used Version V1.5, it is actually running, however, I still got
error, I checked my log file and tree file, I found the states number
of my tree file and log file is not the same. That might be the
problem.

Best regards,
Jason

2020-01-09 13:17 GMT+01:00, Jiexiong Xie <xjx...@gmail.com>:
> Dear Hovhannes,
>
> Thanks a lot, yes, I indeed used version V1.7.1. I will try the earlier
> versions. thanks !
>
> Best regards,
> Jason
>
>> 在 2020年1月9日，上午11:36，HS <hovo...@gmail.com> 写道：
>>
>> 
>> email to beast-users...@googlegroups.com.
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/beast-users/238da7c9-df53-4087-903d-faee75ab9ef1%40googlegroups.com.
>

Jiexiong Xie

unread,
Jan 9, 2020, 12:20:35 PM1/9/20
to beast...@googlegroups.com
Dear Hovhannes,

Thanks a lot, yes, I indeed used version V1.7.1.  I will try the earlier versions. thanks !

Best regards,
Jason

在 2020年1月9日，上午11:36，HS <hovo...@gmail.com> 写道：


To unsubscribe from this group and stop receiving emails from it, send an email to beast-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/beast-users/238da7c9-df53-4087-903d-faee75ab9ef1%40googlegroups.com.

Alefiya Neemuchwala

unread,
Jan 9, 2020, 12:58:11 PM1/9/20
to beast...@googlegroups.com

Dear Hovhannes

Your suggestion of using the logcombiner worked !

However if I increased my bins to 100000 to draw the Bayesian Skyline Plot it got hung again. What is the significance of the number of bins when plotting the Bayesian Skyline Plots.

Regards

Aly

HS

unread,
Jan 10, 2020, 4:36:50 AM1/10/20
to beast-users
Dear Aly,

The default value for bins is 100. What is the reason you increased it so much? If I am not wrong bins relate to the time span. It should be the number of time fragments where Bayesian Skyline is reconstructed.

Best,
Hovhannes

--
You received this message because you are subscribed to the Google Groups "beast-users" group.

To unsubscribe from this group and stop receiving emails from it, send an email to beast...@googlegroups.com.

Andrew Rambaut

unread,
Jan 10, 2020, 5:06:31 AM1/10/20
to beast...@googlegroups.com
The number of bins in the Tracer just affects how it is drawn. It is essentially drawing a histogram. If you set it too high it will look ’noisy’, too low it will be ‘course’. You certainly don’t want more than about 1000 and for most purposes I would think 100 would be fine. 

Andrew

To unsubscribe from this group and stop receiving emails from it, send an email to beast-users...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/beast-users/393c3045-f1b6-407c-abd6-d055eca815e0%40googlegroups.com.


Reply all
Reply to author
Forward
0 new messages