Hello, I have been working with DiffBind for 6 months.  When I updated DiffBind 2.6.1 last week I got some warnings. I don't understand what's wrong. I used  R CMD INSTALL DiffBind_2.6.1.tar.gz. same as I installed DiffBind 2.4.8

Warning messages:
1: no function found corresponding to methods exports from âDelayedArrayâ for: âacbindâ, âarbindâ
2: no function found corresponding to methods exports from âSummarizedExperimentâ for: âacbindâ, âarbindâ
3: multiple methods tables found for ârglistâ
4: multiple methods tables found for âOntologyâ
5: replacing previous import âAnnotationDbi::Ontologyâ by âBiocGenerics::Ontologyâ when loading âGenomicFeaturesâ
6: In .recacheSubclasses(def@className, def, doSubclasses,  ... :
  undefined subclass "GIntervalTree" of class "GenomicRangesORGenomicRangesList"; definition not updated
7: In .recacheSubclasses(def@className, def, doSubclasses,  ... :
  undefined subclass "GIntervalTree" of class "RangedDataORGenomicRanges"; definition not updated
8: In .recacheSubclasses(def@className, def, doSubclasses,  ... :
  undefined subclass "RangedDataList" of class "RangedDataORRangedDataList"; definition not updated
9: multiple methods tables found for âspaceâ
10: replacing previous import âBiocGenerics::Ontologyâ by âAnnotationDbi::Ontologyâ when loading âannotateâ
11: multiple methods tables found for âOntologyâ
12: replacing previous import âBiocGenerics::Ontologyâ by âAnnotationDbi::Ontologyâ when loading âAnnotationForgeâ

Then previously worked scripts failed to run in DiffBind 2.6.1. I reinstalled the older version 2.4.8 again,however, dba.count still failed if I chose score=DBA_SCORE_SUMMIT but works if I chose score=DBA_SCORE_RPKM. Since the peak calling from ATAC-seq , I was told score=DBA_SCORE_SUMMIT is recommended. Could anyone help me with the problem? Thanks a lot.

Error Information
> eiff<-dba.count(eiff, minOverlap=12,score=DBA_SCORE_SUMMIT,summits=150,readFormat=DBA_READS_DEFAULT, bParallel=eiff$config$RunParallel)
Re-centering peaks...
Error in heights * sapply(called, function(x) x) : non-conformable arrays

> sessionInfo()
R version 3.4.2 (2017-09-28)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 14.04.5 LTS

Matrix products: default
BLAS: /usr/lib/openblas-base/libblas.so.3
LAPACK: /usr/lib/lapack/liblapack.so.3.0

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  stats4    stats     graphics  grDevices utils     datasets 
[8] methods   base     

other attached packages:
 [1] DiffBind_2.6.1             SummarizedExperiment_1.6.5
 [3] DelayedArray_0.2.7         matrixStats_0.52.2        
 [5] Biobase_2.36.2             GenomicRanges_1.30.0      
 [7] GenomeInfoDb_1.14.0        IRanges_2.12.0            
 [9] S4Vectors_0.16.0           BiocGenerics_0.24.0       

loaded via a namespace (and not attached):
 [1] edgeR_3.18.1             bit64_0.9-7              splines_3.4.2           
 [4] gtools_3.5.0             assertthat_0.2.0         latticeExtra_0.6-28     
 [7] amap_0.8-14              RBGL_1.52.0              blob_1.1.0              
[10] GenomeInfoDbData_0.99.0  Rsamtools_1.28.0         ggrepel_0.7.0           
[13] Category_2.42.1          RSQLite_2.0              backports_1.1.1         
[16] lattice_0.20-35          glue_1.2.0               limma_3.32.7            
[19] digest_0.6.12            RColorBrewer_1.1-2       XVector_0.16.0          
[22] checkmate_1.8.5          colorspace_1.3-2         Matrix_1.2-12           
[25] plyr_1.8.4               GSEABase_1.38.2          XML_3.98-1.9            
[28] pkgconfig_2.0.1          pheatmap_1.0.8           ShortRead_1.34.1        
[31] biomaRt_2.32.1           genefilter_1.58.1        zlibbioc_1.22.0         
[34] xtable_1.8-2             GO.db_3.4.1              scales_0.5.0            
[37] brew_1.0-6               gdata_2.18.0             BiocParallel_1.10.1     
[40] tibble_1.3.4             annotate_1.54.0          ggplot2_2.2.1           
[43] GenomicFeatures_1.28.5   lazyeval_0.2.0           magrittr_1.5            
[46] survival_2.41-3          memoise_1.1.0            systemPipeR_1.10.2      
[49] fail_1.3                 gplots_3.0.1             hwriter_1.3.2           
[52] GOstats_2.42.0           graph_1.54.0             tools_3.4.2             
[55] BBmisc_1.11              stringr_1.2.0            sendmailR_1.2-1         
[58] munsell_0.4.3            locfit_1.5-9.1           bindrcpp_0.2            
[61] AnnotationDbi_1.38.2     Biostrings_2.44.2        compiler_3.4.2          
[64] caTools_1.17.1           rlang_0.1.4              grid_3.4.2              
[67] RCurl_1.95-4.8           rjson_0.2.15             AnnotationForge_1.18.2  
[70] bitops_1.0-6             base64enc_0.1-3          gtable_0.2.0            
[73] DBI_0.7                  R6_2.2.2                 GenomicAlignments_1.12.2
[76] dplyr_0.7.4              rtracklayer_1.30.4       bit_1.1-12              
[79] bindr_0.1                KernSmooth_2.23-15       stringi_1.1.5           
[82] BatchJobs_1.6            Rcpp_0.12.13

I can't reproduce this with my test data. Is there a reason you can't use biocLite() to update the packages? That will make sure all the dependencies are on the correct version -- I can see some inconsistencies in your version numbers (e.g. SummarizedExperiment, which should be 1.8.x in Bioconductor 3.6).

The good news is that you don't need to set score=DBA_SCORE_SUMMITS. For one thing, this won't actually make any difference to the analysis, as it is not used by dba.analyze() (it is just a score for plots of unanalyzed data, or when exporting the binding matrix). Also, I don't think it would be a good score for ATAC data as it is most useful when working with very sharp peaks, such as a transcription factor with a specific motif. You are better off with a score that captures the read density across the accesschromatin intervals captured by ATAC.

I'm not sure exactly what advice you got, but it was probably to use the summits parameter (as you are doing). By setting summits=150, DiffBind will re-center your peaks around the point of greatest coverage across your samples, and set the peaks to uniform intervals 301bp long (summit + 150bp each side). 

Cheers-

Rory