Using DMRcate::DMR.plot() after finding DMRs with dmrff::dmrff()

[samuel-carleial]

Hi,
after running an EWAS and getting an eBayes(lmFit()) object, I ran the function dmrff and found some DMRs:

	chr	start	end	n	B	S	estimate	se	z	p.value	p.adjust
339	chr21	43982165	43982560	3	-0.0076364	0.0014123	-0.0076364	0.0014123	-5.406970	1e-07	0.0248778
351	chr3	44465202	44465646	3	-0.0054155	0.0008902	-0.0054155	0.0008902	-6.083507	0e+00	0.0004563
355	chr4	84257182	84257282	2	-0.0094295	0.0010579	-0.0094295	0.0010579	-8.912986	0e+00	0.0000000
405	chr8	134306801	134307105	2	-0.0058260	0.0010349	-0.0058260	0.0010349	-5.629692	0e+00	0.0070066

These DMRs could be mapped to some sites:

	region	site	chr	pos	UCSC_RefGene_Name	estimate	p.value	dmr.start	dmr.end	dmr.z	dmr.p.adjust
354807	1	354807	chr21	43982165	SLC37A1	-0.0028612	0.0115771	43982165	43982560	-5.406970	0.0248778
354808	1	354808	chr21	43982270	SLC37A1;SLC37A1	-0.0014308	0.0853719	43982165	43982560	-5.406970	0.0248778
354809	1	354809	chr21	43982560	SLC37A1	-0.0030894	0.0019277	43982165	43982560	-5.406970	0.0248778
68242	2	68242	chr3	44465202		-0.0018421	0.0288792	44465202	44465646	-6.083507	0.0004563
68243	2	68243	chr3	44465567		-0.0036571	0.0003185	44465202	44465646	-6.083507	0.0004563
68244	2	68244	chr3	44465646		-0.0017130	0.0497925	44465202	44465646	-6.083507	0.0004563
93196	3	93196	chr4	84257182	HPSE;HPSE;HPSE;HPSE	-0.0034672	0.0000125	84257182	84257282	-8.912986	0.0000000
93197	3	93197	chr4	84257282	HPSE;HPSE;HPSE;HPSE	-0.0034962	0.0011262	84257182	84257282	-8.912986	0.0000000
179771	4	179771	chr8	134306801	NDRG1;NDRG1	-0.0025517	0.0092720	134306801	134307105	-5.629692	0.0070066
179772	4	179772	chr8	134307105	NDRG1;NDRG1	-0.0018520	0.0364748	134306801	134307105	-5.629692	0.0070066

Now, I would like to simply plot these DMRs using the DMR.plot function, but I am having a hard time including the arguments of the function. I keep receiving the error message: Error in nwrap[!needsWrapping] <- nn[!needsWrapping] : NAs are not allowed in subscripted assignments. The genomic ranges was made manually and look like this:

GRanges object with 4 ranges and 3 metadata columns:
      seqnames              ranges strand |   no.cpgs min_smoothed_fdr overlapping.genes
         <Rle>           <IRanges>  <Rle> | <numeric>        <numeric>       <character>
  [1]    chr21   43982165-43982560      * |         3      2.48778e-02           SLC37A1
  [2]     chr3   44465202-44465646      * |         3      4.56346e-04                  
  [3]     chr4   84257182-84257282      * |         2      1.92799e-13              HPSE
  [4]     chr8 134306801-134307105      * |         2      7.00664e-03             NDRG1
  -------
  seqinfo: 4 sequences from an unspecified genome; no seqlengths

Beta matrix and phenotype trait are referring to the lmFit above. The function I am using is like this:

DMR.plot(ranges=g2, dmr=1, CpGs=beta_vals, what="Beta",
         arraytype = "EPIC", phen.col=pheno_BL$group, genome="hg19")

Any help is appreciated

[AgiaJorda]

Did you ever resolve this error? It is something I am now running into :)

[AgiaJorda]

Update I have found that clearing the workspace, history etc. quitting and restarting R seems to resolve the error temporarily. But the error does return at random.

[perishky]

Could you send me an example that I can try myself? I haven't been able to reproduce the error, but that's likely because I'm doing something different or have different data!