[fraise13]

Error in sqliteExecStatement(con, statement, bind.data) :

Hi All,

I seem to be having similar problems as everyone else and I have looked through many of the posts without finding the solution. However, I am new to R and RNA-seq in general.

Here is what I am getting in RStudio:
> cuff
CuffSet instance with:
4 samples
19856 genes
24928 isoforms
21518 TSS
23593 CDS
119136 promoters
129108 splicing
119136 relCDS
> disp<-dispersionPlot(gene(cuff))
Error in dispersionPlot(gene(cuff)) :
error in evaluating the argument 'object' in selecting a method for function 'dispersionPlot': Error in (function (classes, fdef, mtable) :
unable to find an inherited method for function ‘gene’ for signature ‘"CuffSet"’
> disp<-dispersionPlot(genes(cuff))
> disp
Error in `$<-.data.frame`(`*tmp*`, "SCALE_X", value = 1L) :
replacement has 1 row, data has 0
In addition: Warning message:
In max(panels$ROW) : no non-missing arguments to max; returning -Inf
> genes.scv<-fpkmSCVPlot(genes(cuff))
Error in sqliteExecStatement(con, statement, bind.data) :
RS-DBI driver: (error in statement: near ")": syntax error)
> dens<-csDensity(genes(cuff))
> dens
Warning messages:
1: Removed 1414 rows containing non-finite values (stat_density).
2: Removed 2084 rows containing non-finite values (stat_density).
3: Removed 1457 rows containing non-finite values (stat_density).
4: Removed 1150 rows containing non-finite values (stat_density).

In particular, this error: Error in sqliteExecStatement(con, statement, bind.data) :
RS-DBI driver: (error in statement: near ")": syntax error)
seems to be common with everyone. Has anyone figured out the cause and how to fix it?

I did add the gtfFile and genome arguments as suggested.

I would really appreciate any help. I am using v. 0.97551 of RStudio and Cuffdiff v. 2.1.1 on Galaxy.

Thanks,
Fres

[adiallo]

Hello,
I had the same problem. I think you are using a version of Cufflinks < 2.0.0. And cummeRbund 2.0 needs Cufflinks >= 2.0.0.
So you need to redo all the cuff part with a newer version. Or use cummRBund 1.0.

Cheers ,
Alpha

[fraise13]

Thank you Alpha. I will look into that. I also found that part of the problem may be that cummeRbund requires some files (read_groups.info) that cuffdiff in Galaxy does not include in the output. Galaxy may add an option to include these files in the future.

[IBseq]

Dear all,
any suggestion about this issue I have?

I run the following commands:
> library (cummeRbund)
Loading required package: RSQLite
Loading required package: DBI
Loading required package: ggplot2
Loading required package: reshape2
> cuff<-readCufflinks ()
> cuff
CuffSet instance with:
2 samples
24302 genes
42262 isoforms
28828 TSS
25037 CDS
24302 promoters
28828 splicing
0 relCDS
> iso.diff <- diffData(isoforms(cuff))
> iso.diff.top <- iso.diff[order(iso.diff$q_value),][1:20,]
> iso.diff.top
isoform_id sample_1 sample_2 status value_1 value_2
19598 TCONS_00019647 q1 q2 OK 0.00142518 63.8030
23003 TCONS_00023056 q1 q2 OK 0.01873000 267.5380
16085 TCONS_00016126 q1 q2 OK 0.00828880 53.7744
20375 TCONS_00020425 q1 q2 OK 0.01018440 41.0838
14718 TCONS_00014757 q1 q2 OK 0.01255110 52.7526
38909 TCONS_00038994 q1 q2 OK 0.00506554 18.4579
9364 TCONS_00009383 q1 q2 OK 0.01624700 41.5996
21567 TCONS_00021617 q1 q2 OK 0.02047770 43.5210
21617 TCONS_00021668 q1 q2 OK 0.07737150 173.5400
12908 TCONS_00012945 q1 q2 OK 0.01454510 29.1976
2916 TCONS_00002919 q1 q2 OK 0.02808040 52.5176
11888 TCONS_00011918 q1 q2 OK 0.01283370 23.4124
24967 TCONS_00025024 q1 q2 OK 0.07871780 93.7237
36578 TCONS_00036659 q1 q2 OK 0.15108000 179.3100
18827 TCONS_00018874 q1 q2 OK 0.19685300 402.2040
26748 TCONS_00026809 q1 q2 OK 0.07271350 66.3874
24573 TCONS_00024628 q1 q2 OK 0.01203420 20.7750
5805 TCONS_00005818 q1 q2 OK 0.00198928 2.9240
35808 TCONS_00035884 q1 q2 OK 0.10710700 108.1710
26864 TCONS_00026926 q1 q2 OK 0.05353970 86.0766
log2_fold_change test_stat p_value q_value significant
19598 15.45020 -7.91793 2.44249e-15 5.34124e-11 yes
23003 13.80210 -7.39372 1.42775e-13 1.56110e-09 yes
16085 12.66350 -7.33093 2.28484e-13 1.66550e-09 yes
20375 11.97800 -7.27264 3.52607e-13 1.92770e-09 yes
14718 12.03720 -7.23914 4.51417e-13 1.97432e-09 yes
38909 11.83120 -7.17187 7.39853e-13 2.69652e-09 yes
9364 11.32220 -6.84004 7.91722e-12 2.47334e-08 yes
21567 11.05340 -6.73898 1.59504e-11 3.87558e-08 yes
21617 11.13120 -6.75063 1.47207e-11 3.87558e-08 yes
12908 10.97110 -6.67956 2.39653e-11 5.24073e-08 yes
2916 10.86900 -6.61070 3.82498e-11 6.62383e-08 yes
11888 10.83310 -6.59542 4.24061e-11 6.62383e-08 yes
24967 10.21750 -6.60852 3.88183e-11 6.62383e-08 yes
36578 10.21290 -6.59690 4.19833e-11 6.62383e-08 yes
18827 10.99660 -6.54690 5.87441e-11 8.56411e-08 yes
26748 9.83447 -6.44433 1.16116e-10 1.58701e-07 yes
24573 10.75350 -6.42860 1.28785e-10 1.65663e-07 yes
5805 10.52150 -6.40323 1.52120e-10 1.84809e-07 yes
35808 9.98005 -6.39398 1.61620e-10 1.86017e-07 yes
26864 10.65080 -6.35030 2.14901e-10 2.34972e-07 yes


however, these top 20 include transcripts that start with NM_00000 and NR_00000

what I want is to select only the NM_ as these are proteins/transcript, NR_ if I am not wrong is the gene (here cummerbund only gives TCONS, but I have double checked themn with the NM/NR nomenclature and I know they are mixed up as it is listing by fold change...)

How do I command this on cummerbund?

thanks,
ibseq

[tokitky]

Hi all,

I have a similar problem.

>genediffID <- read.table("genediffID.txt",header= F, sep="/t")
> myGenes <-getGenes(cuff,genediffID)
Error in sqliteExecStatement(conn, statement, ...) :
RS-DBI driver: (RS_SQLite_exec: could not execute1: cannot start a transaction within a transaction)

genediffID was orgninally a column vector but row vector format seemed to be appropriate.

I solved the problem as follows.

>gendiffID <- t(genediffID)
> myGenes <-getGenes(cuff,genediffID)

I also had a similar diffrent problem.

>genediffID2 <- read.table("genediffID2.txt",header= F, sep="/t")
> myGenes2 <-getGenes(cuff,genediffID2)

Error in sqliteExecStatement(conn, statement, bind.data) :
RS-DBI driver: (expired SQLiteConnection)

At this time, cuff was broken by some reason.
So I made cuff again.

>cuff <- readCufflinks("input folder")
>genediffID2 <- read.table("genediffID2.txt",header= F, sep="/t")
> myGenes2 <-getGenes(cuff,genediffID2)

then, it worked.

[IBseq]

dear cummerbund users,
i am trying to run the following command:

library(cummeRbund)
cuff <- readCufflinks()
....

iso.diff <- diffData(isoforms(cuff))
iso.diff.top <- iso.diff[order(iso.diff$q_value),][1:20,]

which works fine, however I would like to run it not through all my samples but a list I have selected (a family of proteins i have extracted and only the significant ones).

I made this table with the isoforms of interest but I don't know how to input the command....any suggestions? I saved this table where all the cuffdiff outputs are and called it myisoforms_exp.diff (thus similar to the original file)

I did not understand how we create a gene set, so I thought this way might work..


thanks,
ibseq

[nilesh.gardi2688]

Heatmap in cummeRbund package

How to alter the scale of heatmap in cummerBund? heatMidpoint option in csHeatmap is seems to be non functional.

[adiallo]

Hello
Use :
myGeneIds <- scan("C://heat//tcf4_ensg.txt",what="character")
myGenes<-getGenes(cuff,myGeneIds)

h<-csHeatmap(myGenes,cluster='both')

Cheers

[nancysky]

Same error as yours

Hi Fres,

I'm having exactly the same error right now. Can I ask how did you solve the problem? Finding a higher version of cufflinks? How could you find it? I only see one version of cufflinks provided by the server, and that one is older. I can see I have something in cuff but just cannot let the anticipated disp.

Thanks a lot!!!
Nancy

Hi All,

I seem to be having similar problems as everyone else and I have looked through many of the posts without finding the solution. However, I am new to R and RNA-seq in general.

Here is what I am getting in RStudio:
> cuff
CuffSet instance with:
4 samples
19856 genes
24928 isoforms
21518 TSS
23593 CDS
119136 promoters
129108 splicing
119136 relCDS
> disp<-dispersionPlot(gene(cuff))
Error in dispersionPlot(gene(cuff)) :
error in evaluating the argument 'object' in selecting a method for function 'dispersionPlot': Error in (function (classes, fdef, mtable) :
unable to find an inherited method for function ‘gene’ for signature ‘"CuffSet"’
> disp<-dispersionPlot(genes(cuff))
> disp
Error in `$<-.data.frame`(`*tmp*`, "SCALE_X", value = 1L) :
replacement has 1 row, data has 0
In addition: Warning message:
In max(panels$ROW) : no non-missing arguments to max; returning -Inf
> genes.scv<-fpkmSCVPlot(genes(cuff))
Error in sqliteExecStatement(con, statement, bind.data) :
RS-DBI driver: (error in statement: near ")": syntax error)
> dens<-csDensity(genes(cuff))
> dens
Warning messages:
1: Removed 1414 rows containing non-finite values (stat_density).
2: Removed 2084 rows containing non-finite values (stat_density).
3: Removed 1457 rows containing non-finite values (stat_density).
4: Removed 1150 rows containing non-finite values (stat_density).

In particular, this error: Error in sqliteExecStatement(con, statement, bind.data) :
RS-DBI driver: (error in statement: near ")": syntax error)
seems to be common with everyone. Has anyone figured out the cause and how to fix it?

I did add the gtfFile and genome arguments as suggested.

I would really appreciate any help. I am using v. 0.97551 of RStudio and Cuffdiff v. 2.1.1 on Galaxy.

Thanks,
Fres

[Gonza]

Hi all,

i am getting this error and cannot figure put how to fix it. Please let me know if you have some tips.
Thans

cuff<-readCufflinks()
cuff

Error in sqliteExecStatement(con, statement, bind.data) :
RS-DBI driver: (error in statement: database disk image is malformed)

[adiallo]

Hello Gonza,
What version of the cufflinks package did you use.
Did all the steps went well, without mistakes until the cuffdiff command ?

Here is an example of code that works for me:
library(cummeRbund)

cuff_71_69 <- readCufflinks('_cuffdiff_results_path',rebuild=T)

#===============================================================================================================================================
# To assess the distributions of FPKM scores across samples, you can use the csDensity plot.
dens<-csDensity(genes(cuff_71_69))
dens

dens<-csDensity(genes(cuff_71_69),logMode=TRUE, pseudocount=0.0001)
dens


#===============================================================================================================================================
# Boxplots can be visualized using the csBoxplot method.
b<-csBoxplot(genes(cuff_71_69))
b

Cheers,
Alpha

[Gonza]

Thanks for your reply.
I used cufflinks v2.2.1 and all the steps went well (in fact, in some else computer I was able to upload cuffdiff to R, for some reason in mine does not work - and i have the latest R installed).
Will try your code though.
G

[skdhanraj]

Hi CummeRbund users,

This might be simple question but can anybody explain these warning messages after readCufflinks() command:


Warning messages:
1: attributes are not identical across measure variables; they will be dropped
.
.
8: attributes are not identical across measure variables; they will be dropped

Are these attributes from merged.gtf file?

Thanks,
SK

[SrCardgage]

Warning messages:
1: attributes are not identical across measure variables; they will be dropped
.
.
8: attributes are not identical across measure variables; they will be dropped

Are these attributes from merged.gtf file?

Thanks,
SK

I am seeing these warnings as well. Has some file not been created correctly?

I am not looking for novel genes/transcripts. So, I went directly from tophat (with --no-novel-juncs flag) to cuffdiff. I doubt this would make a difference, but cannot be sure.

[Gonza]

Hi All,

After a lot of struggle I finally found out what the problem was. In my case, the genes.gtf filed used had some # symbols. When importing cuffdiff to R, the lines containing # made it fail. I created a new genes.gtf without # and it worked.

$sed -r 's/#//g' genes.gtf > new_genes.gtf

That it was my problem, I do not know how 'universal' of a problem that is. But it may be a good idea to check your gtf file.

######################################################################################################

BTW, I am assuming you already did the upgrade:

>source("http://www.bioconductor.org/biocLite.R")
>biocLite("BiocUpgrade")

Then, install "cummeRbund"
>biocLite("cummeRbund")

Last, install or reinstall "RSQLite"
>biocLite("RSQLite")