eblancoga / spikchip Goto Github PK

I was trying to apply the spikChIP methodology to my data and the script ended abruptly at this point:

%%%% Run local regression adjusting sample values with spike values
%%%% Performing the analysis on average values
%%%% R CMD BATCH Rscripts/1-2-3-4-5-6-7-8-9-10-11-12-13-14-15-16_10000_avg.R
%%%% R running: Error in read.table("results/1-2-3-4-5-6-7-8-9-10-11-12-13-14-15-16_SPIKCHIP_10000_spike-sample_avg_names.txt") :

%%%% [Wed Aug 23 09:54:54 2023] Stage X. Abruptly finishing the work
%%%% Total running time (hours): 0.578 hours
%%%% Total running time (minutes): 34.683 mins
%%%% Total running time (seconds): 2081 secs [OK]

I've looked to the table "results/1-2-3-4-5-6-7-8-9-10-11-12-13-14-15-16_SPIKCHIP_10000_spike-sample_avg_names.txt" and it is empty.
Can you help me with this error? Thanks!

Hi,
Thank you for this nice tool, I recently tested it on some Chip-RX data from my lab.
I would like to ask for your help about a couple of issues I faced.

Among the output files, I get the 1000_avg_normalized_sample only for the SPIKCHIP method. I converted this file to bedGraph and then to bigwig. However, I would like to create a bigwig also for the other normalization methods in order to compare them. How could I achieve this?

Moreover, when I open in igv the output bigwigs files, for two samples of the same histone modification I get a high signal in the background regions that is also constant. I am struggling to understand what is happening.

See here:

Sample 1 and 3 are from the same histone modification (H3K27Me3), whereas 2 and 4 are from H2Ak119ub1

Hope you can help me out.
Thanks!
Best

Davide

Hi,

First of all, thank you for making this software.

I would like to use a bin size of 20 bp (or 50bp), because I would like higher ChIP-seq signal resolution.
Qn 1: Would I lose reliability of normalisation if I use bin sizes this small?

You mention "selection of this [1000 bp] was based on finding a balance between ChIP-seq resolution, and the computational memory demand and running time required for a more compact bin segmentation".
Qn 2: Could I get an idea of how memory demand and run time scales for smaller bin sizes please?

Thank you,
PGS

Hi,

Thank you for developing such a useful tool. I'm currently attempting to use spikChIP with human cell line samples and have encountered an error while running the R script (avg.R). The error message is:

mn <- normalize.loess(m,subset=s)
Error in simpleLoess(y, x, w, span, degree = degree, parametric = parametric,  : 
  NA/NaN/Inf in foreign function call (arg 1)
In addition: There were 44 warnings (use warnings() to see them)

All values within the script are greater than 0.1, and there are no NAs, at least.

Here is my R session:

sessionInfo()
R version 4.0.3 (2020-10-10)
Platform: x86_64-conda-linux-gnu (64-bit)
Running under: Ubuntu 20.04.6 LTS

Matrix products: default

locale:
 [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
 [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
 [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
 [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C            
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] parallel  stats     graphics  grDevices utils     datasets  methods  
[8] base     

other attached packages:
[1] MASS_7.3-60.0.1     affy_1.68.0         Biobase_2.50.0     
[4] BiocGenerics_0.36.1

loaded via a namespace (and not attached):
[1] BiocManager_1.30.22   zlibbioc_1.36.0       compiler_4.0.3       
[4] affyio_1.60.0         preprocessCore_1.52.1

Could you provide some guidance on resolving this issue? I appreciate your help.

Thanks