Problem with nearest() in GenomicRanges

0

Entering edit mode

Chris Whelan ▴ 60

@chris-whelan-4779

Last seen 10.2 years ago

Hi all, I'm having an issue with the nearest() method in GenomicRanges. I've created GRanges from two data frames, from data for a species genome assembly with lots of scaffolds: > xs <- with(x.data, GRanges(seqnames=Scaffold, ranges=IRanges(start=Start, end=Stop), strand=Strand)) > xs GRanges with 989 ranges and 0 elementMetadata cols: seqnames ranges strand <rle> <iranges> <rle> [1] GL397480 [ 1477346, 1477813] - [2] GL397446 [ 869823, 870295] + ... ... ... ... [988] GL397464 [1138154, 1138427] - [989] GL397464 [1138154, 1138427] - --- seqlengths: ADFV01135143 ADFV01136404 GL397262 ... GL397832 GL397836 NA NA NA ... NA NA > cpg.islands <- with(cpg.island.data, GRanges(seqnames=V1, ranges=IRanges(start=V4, end=V5), strand='*')) > cpg.islands GRanges with 88968 ranges and 0 elementMetadata cols: seqnames ranges strand <rle> <iranges> <rle> [1] GL397261 [ 4409, 5062] * [2] GL397261 [ 66765, 67010] * ... ... ... ... [88967] ADFV01162706 [1834, 2052] * [88968] ADFV01162751 [ 226, 884] * --- seqlengths: ADFV01134425 ADFV01134436 ADFV01134442 ... GL405215 GL405216 NA NA NA ... NA NA And I'd like to use nearest to find the closest island to each x: > nearest(xs, cpg.islands) Error in Rle(values = callGeneric(runValue(e1)[which1], runValue(e2)[which2]), : error in evaluating the argument 'values' in selecting a method for function 'Rle': Error in Ops.factor(runValue(e1)[which1], runValue(e2)[which2]) : level sets of factors are different > traceback() 12: Rle(values = callGeneric(runValue(e1)[which1], runValue(e2)[which2]), lengths = diffWithInitialZero(ends), check = FALSE) 11: seqnames(x) != seqnames(y) 10: seqnames(x) != seqnames(y) 9: `seqselect<-`(`*tmp*`, seqnames(x) != seqnames(y), value = NA) 8: `seqselect<-`(`*tmp*`, seqnames(x) != seqnames(y), value = NA) 7: .local(x, y, ...) 6: distance(x[midx], subject[p[midx]]) 5: distance(x[midx], subject[p[midx]]) 4: .nearest(x, subject, ignore.strand, ...) 3: .local(x, subject, ...) 2: nearest(xs, cpg.islands) 1: nearest(xs, cpg.islands) This same data works fine with other methods like findOverlaps(): > findOverlaps(xs, cpg.islands) Hits of length 225 queryLength: 989 subjectLength: 88968 queryHits subjectHits <integer> <integer> 1 2 75585 2 6 36857 ... ... ... 224 963 84872 225 964 84872 Any idea what I might be doing wrong? Here's my sessionInfo(): > sessionInfo() R version 2.15.1 (2012-06-22) Platform: x86_64-pc-linux-gnu (64-bit) locale: [1] LC_CTYPE=C LC_NUMERIC=C [3] LC_TIME=en_US.UTF-8 LC_COLLATE=C [5] LC_MONETARY=C LC_MESSAGES=C [7] LC_PAPER=C LC_NAME=C [9] LC_ADDRESS=C LC_TELEPHONE=C [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] locfit_1.5-8 GenomicFeatures_1.8.2 AnnotationDbi_1.18.1 [4] Biobase_2.16.0 multicore_0.1-7 rtracklayer_1.16.3 [7] GenomicRanges_1.8.9 IRanges_1.14.4 BiocGenerics_0.2.0 [10] reshape2_1.2.1 ggplot2_0.9.1 BiocInstaller_1.4.7 loaded via a namespace (and not attached): [1] BSgenome_1.24.0 Biostrings_2.24.1 DBI_0.2-5 MASS_7.3-19 [5] RColorBrewer_1.0-5 RCurl_1.91-1 RSQLite_0.11.1 Rsamtools_1.8.5 [9] XML_3.9-4 biomaRt_2.12.0 bitops_1.0-4.1 colorspace_1.1-1 [13] compiler_2.15.1 dichromat_1.2-4 digest_0.5.2 grid_2.15.1 [17] labeling_0.1 lattice_0.20-6 memoise_0.1 munsell_0.3 [21] plyr_1.7.1 proto_0.3-9.2 scales_0.2.1 stats4_2.15.1 [25] stringr_0.6.1 tools_2.15.1 zlibbioc_1.2.0 Thanks in advance, Chris [[alternative HTML version deleted]]

GenomicRanges GenomicRanges • 1.4k views

ADD COMMENT • link updated 12.3 years ago by Valerie Obenchain ★ 6.8k • written 12.3 years ago by Chris Whelan ▴ 60

0

Entering edit mode

Valerie Obenchain ★ 6.8k

@valerie-obenchain-4275

Last seen 2.9 years ago

United States

Hi Chris, It's difficult to know what's going on without a reproducible example. Can you attach a small subset of your GRanges that still triggers the error? If not, is there a place I can get the data for testing? Valerie On 07/28/12 12:03, Chris Whelan wrote: > Hi all, > > I'm having an issue with the nearest() method in GenomicRanges. I've > created GRanges from two data frames, from data for a species genome > assembly with lots of scaffolds: > >> xs<- with(x.data, GRanges(seqnames=Scaffold, ranges=IRanges(start=Start, end=Stop), strand=Strand)) >> xs > GRanges with 989 ranges and 0 elementMetadata cols: > seqnames ranges strand > <rle> <iranges> <rle> > [1] GL397480 [ 1477346, 1477813] - > [2] GL397446 [ 869823, 870295] + > ... ... ... ... > [988] GL397464 [1138154, 1138427] - > [989] GL397464 [1138154, 1138427] - > --- > seqlengths: > ADFV01135143 ADFV01136404 GL397262 ... GL397832 GL397836 > NA NA NA ... NA NA > >> cpg.islands<- with(cpg.island.data, GRanges(seqnames=V1, ranges=IRanges(start=V4, end=V5), strand='*')) >> cpg.islands > GRanges with 88968 ranges and 0 elementMetadata cols: > seqnames ranges strand > <rle> <iranges> <rle> > [1] GL397261 [ 4409, 5062] * > [2] GL397261 [ 66765, 67010] * > ... ... ... ... > [88967] ADFV01162706 [1834, 2052] * > [88968] ADFV01162751 [ 226, 884] * > --- > seqlengths: > ADFV01134425 ADFV01134436 ADFV01134442 ... GL405215 GL405216 > NA NA NA ... NA NA > > And I'd like to use nearest to find the closest island to each x: > >> nearest(xs, cpg.islands) > Error in Rle(values = callGeneric(runValue(e1)[which1], > runValue(e2)[which2]), : > error in evaluating the argument 'values' in selecting a method for > function 'Rle': Error in Ops.factor(runValue(e1)[which1], > runValue(e2)[which2]) : > level sets of factors are different >> traceback() > 12: Rle(values = callGeneric(runValue(e1)[which1], runValue(e2)[which2]), > lengths = diffWithInitialZero(ends), check = FALSE) > 11: seqnames(x) != seqnames(y) > 10: seqnames(x) != seqnames(y) > 9: `seqselect<-`(`*tmp*`, seqnames(x) != seqnames(y), value = NA) > 8: `seqselect<-`(`*tmp*`, seqnames(x) != seqnames(y), value = NA) > 7: .local(x, y, ...) > 6: distance(x[midx], subject[p[midx]]) > 5: distance(x[midx], subject[p[midx]]) > 4: .nearest(x, subject, ignore.strand, ...) > 3: .local(x, subject, ...) > 2: nearest(xs, cpg.islands) > 1: nearest(xs, cpg.islands) > > This same data works fine with other methods like findOverlaps(): > >> findOverlaps(xs, cpg.islands) > Hits of length 225 > queryLength: 989 > subjectLength: 88968 > queryHits subjectHits > <integer> <integer> > 1 2 75585 > 2 6 36857 > ... ... ... > 224 963 84872 > 225 964 84872 > > Any idea what I might be doing wrong? Here's my sessionInfo(): > >> sessionInfo() > R version 2.15.1 (2012-06-22) > Platform: x86_64-pc-linux-gnu (64-bit) > > locale: > [1] LC_CTYPE=C LC_NUMERIC=C > [3] LC_TIME=en_US.UTF-8 LC_COLLATE=C > [5] LC_MONETARY=C LC_MESSAGES=C > [7] LC_PAPER=C LC_NAME=C > [9] LC_ADDRESS=C LC_TELEPHONE=C > [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C > > attached base packages: > [1] stats graphics grDevices utils datasets methods base > > other attached packages: > [1] locfit_1.5-8 GenomicFeatures_1.8.2 AnnotationDbi_1.18.1 > [4] Biobase_2.16.0 multicore_0.1-7 rtracklayer_1.16.3 > [7] GenomicRanges_1.8.9 IRanges_1.14.4 BiocGenerics_0.2.0 > [10] reshape2_1.2.1 ggplot2_0.9.1 BiocInstaller_1.4.7 > > loaded via a namespace (and not attached): > [1] BSgenome_1.24.0 Biostrings_2.24.1 DBI_0.2-5 MASS_7.3-19 > [5] RColorBrewer_1.0-5 RCurl_1.91-1 RSQLite_0.11.1 Rsamtools_1.8.5 > [9] XML_3.9-4 biomaRt_2.12.0 bitops_1.0-4.1 colorspace_1.1-1 > [13] compiler_2.15.1 dichromat_1.2-4 digest_0.5.2 grid_2.15.1 > [17] labeling_0.1 lattice_0.20-6 memoise_0.1 munsell_0.3 > [21] plyr_1.7.1 proto_0.3-9.2 scales_0.2.1 stats4_2.15.1 > [25] stringr_0.6.1 tools_2.15.1 zlibbioc_1.2.0 > > Thanks in advance, > > Chris > > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor

ADD COMMENT • link 12.3 years ago Valerie Obenchain ★ 6.8k

0

Entering edit mode

Hi Valerie, I get this behavior using only a small subset of my data - attached. I am loading it like this: x.data <- read.table("x_data.txt", header=T, sep="\t") cpg.island.data <- read.table('cpg_islands_head.gff') In doing some more testing, it seems like I get this error if not all of the seqnames in the query GRanges are present in the subject GRanges. Is that data not supported? I tried setting the levels with: seqlevels(xs) <- seqlevels(cpg.islands) But then I just get back a vector of NA's from nearest(). Thanks, Chris On Jul 30, 2012, at 7:08 PM, Valerie Obenchain wrote: > Hi Chris, > > It's difficult to know what's going on without a reproducible example. > Can you attach a small subset of your GRanges that still triggers the > error? If not, is there a place I can get the data for testing? > > Valerie > > On 07/28/12 12:03, Chris Whelan wrote: >> Hi all, >> >> I'm having an issue with the nearest() method in GenomicRanges. I've >> created GRanges from two data frames, from data for a species genome >> assembly with lots of scaffolds: >> >>> xs<- with(x.data, GRanges(seqnames=Scaffold, ranges=IRanges(start=Start, end=Stop), strand=Strand)) >>> xs >> GRanges with 989 ranges and 0 elementMetadata cols: >> seqnames ranges strand >> <rle> <iranges> <rle> >> [1] GL397480 [ 1477346, 1477813] - >> [2] GL397446 [ 869823, 870295] + >> ... ... ... ... >> [988] GL397464 [1138154, 1138427] - >> [989] GL397464 [1138154, 1138427] - >> --- >> seqlengths: >> ADFV01135143 ADFV01136404 GL397262 ... GL397832 GL397836 >> NA NA NA ... NA NA >> >>> cpg.islands<- with(cpg.island.data, GRanges(seqnames=V1, ranges=IRanges(start=V4, end=V5), strand='*')) >>> cpg.islands >> GRanges with 88968 ranges and 0 elementMetadata cols: >> seqnames ranges strand >> <rle> <iranges> <rle> >> [1] GL397261 [ 4409, 5062] * >> [2] GL397261 [ 66765, 67010] * >> ... ... ... ... >> [88967] ADFV01162706 [1834, 2052] * >> [88968] ADFV01162751 [ 226, 884] * >> --- >> seqlengths: >> ADFV01134425 ADFV01134436 ADFV01134442 ... GL405215 GL405216 >> NA NA NA ... NA NA >> >> And I'd like to use nearest to find the closest island to each x: >> >>> nearest(xs, cpg.islands) >> Error in Rle(values = callGeneric(runValue(e1)[which1], >> runValue(e2)[which2]), : >> error in evaluating the argument 'values' in selecting a method for >> function 'Rle': Error in Ops.factor(runValue(e1)[which1], >> runValue(e2)[which2]) : >> level sets of factors are different >>> traceback() >> 12: Rle(values = callGeneric(runValue(e1)[which1], runValue(e2)[which2]), >> lengths = diffWithInitialZero(ends), check = FALSE) >> 11: seqnames(x) != seqnames(y) >> 10: seqnames(x) != seqnames(y) >> 9: `seqselect<-`(`*tmp*`, seqnames(x) != seqnames(y), value = NA) >> 8: `seqselect<-`(`*tmp*`, seqnames(x) != seqnames(y), value = NA) >> 7: .local(x, y, ...) >> 6: distance(x[midx], subject[p[midx]]) >> 5: distance(x[midx], subject[p[midx]]) >> 4: .nearest(x, subject, ignore.strand, ...) >> 3: .local(x, subject, ...) >> 2: nearest(xs, cpg.islands) >> 1: nearest(xs, cpg.islands) >> >> This same data works fine with other methods like findOverlaps(): >> >>> findOverlaps(xs, cpg.islands) >> Hits of length 225 >> queryLength: 989 >> subjectLength: 88968 >> queryHits subjectHits >> <integer> <integer> >> 1 2 75585 >> 2 6 36857 >> ... ... ... >> 224 963 84872 >> 225 964 84872 >> >> Any idea what I might be doing wrong? Here's my sessionInfo(): >> >>> sessionInfo() >> R version 2.15.1 (2012-06-22) >> Platform: x86_64-pc-linux-gnu (64-bit) >> >> locale: >> [1] LC_CTYPE=C LC_NUMERIC=C >> [3] LC_TIME=en_US.UTF-8 LC_COLLATE=C >> [5] LC_MONETARY=C LC_MESSAGES=C >> [7] LC_PAPER=C LC_NAME=C >> [9] LC_ADDRESS=C LC_TELEPHONE=C >> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C >> >> attached base packages: >> [1] stats graphics grDevices utils datasets methods base >> >> other attached packages: >> [1] locfit_1.5-8 GenomicFeatures_1.8.2 AnnotationDbi_1.18.1 >> [4] Biobase_2.16.0 multicore_0.1-7 rtracklayer_1.16.3 >> [7] GenomicRanges_1.8.9 IRanges_1.14.4 BiocGenerics_0.2.0 >> [10] reshape2_1.2.1 ggplot2_0.9.1 BiocInstaller_1.4.7 >> >> loaded via a namespace (and not attached): >> [1] BSgenome_1.24.0 Biostrings_2.24.1 DBI_0.2-5 MASS_7.3-19 >> [5] RColorBrewer_1.0-5 RCurl_1.91-1 RSQLite_0.11.1 Rsamtools_1.8.5 >> [9] XML_3.9-4 biomaRt_2.12.0 bitops_1.0-4.1 colorspace_1.1-1 >> [13] compiler_2.15.1 dichromat_1.2-4 digest_0.5.2 grid_2.15.1 >> [17] labeling_0.1 lattice_0.20-6 memoise_0.1 munsell_0.3 >> [21] plyr_1.7.1 proto_0.3-9.2 scales_0.2.1 stats4_2.15.1 >> [25] stringr_0.6.1 tools_2.15.1 zlibbioc_1.2.0 >> >> Thanks in advance, >> >> Chris >> >> [[alternative HTML version deleted]] >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor at r-project.org >> https://stat.ethz.ch/mailman/listinfo/bioconductor >> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor >

ADD REPLY • link 12.3 years ago Chris Whelan ▴ 60

0

Entering edit mode

Thanks for sending the test data. This is now fixed in GenomicRanges release 1.8.10 and was fixed in devel a few months ago. The 1.8.10 version should be available through biocLite() ~9am PST Wednesday. The problem was that distance(), which is called from within nearest(), was not handling the case of no matching seqnames. > x.data <- read.table("x.csv", header=TRUE, sep="\t") > xs <- with(x.data, GRanges(seqnames=Scaffold, + ranges=IRanges(start=Start,end=Stop), strand=Strand)) > > cpg.island.data <- read.table('cpg_islands_head.gff') > cpg.islands <- with(cpg.island.data, GRanges(seqnames=V1, + ranges=IRanges(start=V4, end=V5), strand='*')) In these data, none of the seqnames in xs match those in cpg.islands, > runValue(seqnames(xs)) %in% runValue(seqnames(cpg.islands)) [1] FALSE FALSE FALSE Because we have no common seqnames we get a vector of NAs as the result. > nearest(xs, cpg.islands) [1] NA NA NA NA NA NA NA NA NA If there were matching seqnames you would see results for those values. For example, if we set the seqnames in cpg.islands to match one seqname in xs, > seqlevels(cpg.islands) <- "GL397446" > nearest(xs, cpg.islands) [1] NA 10 10 10 NA NA NA NA NA Valerie On 07/30/2012 08:17 PM, Chris Whelan wrote: > Hi Valerie, > > I get this behavior using only a small subset of my data - attached. I > am loading it like this: > > x.data <- read.table("x_data.txt", header=T, sep="\t") > cpg.island.data <- read.table('cpg_islands_head.gff') > > In doing some more testing, it seems like I get this error if not all > of the seqnames in the query GRanges are present in the subject > GRanges. Is that data not supported? > > I tried setting the levels with: > > seqlevels(xs) <- seqlevels(cpg.islands) > > But then I just get back a vector of NA's from nearest(). > > Thanks, > > Chris > > > > > On Jul 30, 2012, at 7:08 PM, Valerie Obenchain wrote: > > > Hi Chris, > > > > It's difficult to know what's going on without a reproducible example. > > Can you attach a small subset of your GRanges that still triggers the > > error? If not, is there a place I can get the data for testing? > > > > Valerie > > > > On 07/28/12 12:03, Chris Whelan wrote: > >> Hi all, > >> > >> I'm having an issue with the nearest() method in GenomicRanges. I've > >> created GRanges from two data frames, from data for a species genome > >> assembly with lots of scaffolds: > >> > >>> xs<- with(x.data, GRanges(seqnames=Scaffold, > ranges=IRanges(start=Start, end=Stop), strand=Strand)) > >>> xs > >> GRanges with 989 ranges and 0 elementMetadata cols: > >> seqnames ranges strand > >> <rle> <iranges> <rle> > >> [1] GL397480 [ 1477346, 1477813] - > >> [2] GL397446 [ 869823, 870295] + > >> ... ... ... ... > >> [988] GL397464 [1138154, 1138427] - > >> [989] GL397464 [1138154, 1138427] - > >> --- > >> seqlengths: > >> ADFV01135143 ADFV01136404 GL397262 ... GL397832 GL397836 > >> NA NA NA ... NA NA > >> > >>> cpg.islands<- with(cpg.island.data, GRanges(seqnames=V1, > ranges=IRanges(start=V4, end=V5), strand='*')) > >>> cpg.islands > >> GRanges with 88968 ranges and 0 elementMetadata cols: > >> seqnames ranges strand > >> <rle> <iranges> <rle> > >> [1] GL397261 [ 4409, 5062] * > >> [2] GL397261 [ 66765, 67010] * > >> ... ... ... ... > >> [88967] ADFV01162706 [1834, 2052] * > >> [88968] ADFV01162751 [ 226, 884] * > >> --- > >> seqlengths: > >> ADFV01134425 ADFV01134436 ADFV01134442 ... GL405215 GL405216 > >> NA NA NA ... NA NA > >> > >> And I'd like to use nearest to find the closest island to each x: > >> > >>> nearest(xs, cpg.islands) > >> Error in Rle(values = callGeneric(runValue(e1)[which1], > >> runValue(e2)[which2]), : > >> error in evaluating the argument 'values' in selecting a method for > >> function 'Rle': Error in Ops.factor(runValue(e1)[which1], > >> runValue(e2)[which2]) : > >> level sets of factors are different > >>> traceback() > >> 12: Rle(values = callGeneric(runValue(e1)[which1], > runValue(e2)[which2]), > >> lengths = diffWithInitialZero(ends), check = FALSE) > >> 11: seqnames(x) != seqnames(y) > >> 10: seqnames(x) != seqnames(y) > >> 9: `seqselect<-`(`*tmp*`, seqnames(x) != seqnames(y), value = NA) > >> 8: `seqselect<-`(`*tmp*`, seqnames(x) != seqnames(y), value = NA) > >> 7: .local(x, y, ...) > >> 6: distance(x[midx], subject[p[midx]]) > >> 5: distance(x[midx], subject[p[midx]]) > >> 4: .nearest(x, subject, ignore.strand, ...) > >> 3: .local(x, subject, ...) > >> 2: nearest(xs, cpg.islands) > >> 1: nearest(xs, cpg.islands) > >> > >> This same data works fine with other methods like findOverlaps(): > >> > >>> findOverlaps(xs, cpg.islands) > >> Hits of length 225 > >> queryLength: 989 > >> subjectLength: 88968 > >> queryHits subjectHits > >> <integer> <integer> > >> 1 2 75585 > >> 2 6 36857 > >> ... ... ... > >> 224 963 84872 > >> 225 964 84872 > >> > >> Any idea what I might be doing wrong? Here's my sessionInfo(): > >> > >>> sessionInfo() > >> R version 2.15.1 (2012-06-22) > >> Platform: x86_64-pc-linux-gnu (64-bit) > >> > >> locale: > >> [1] LC_CTYPE=C LC_NUMERIC=C > >> [3] LC_TIME=en_US.UTF-8 LC_COLLATE=C > >> [5] LC_MONETARY=C LC_MESSAGES=C > >> [7] LC_PAPER=C LC_NAME=C > >> [9] LC_ADDRESS=C LC_TELEPHONE=C > >> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C > >> > >> attached base packages: > >> [1] stats graphics grDevices utils datasets methods base > >> > >> other attached packages: > >> [1] locfit_1.5-8 GenomicFeatures_1.8.2 AnnotationDbi_1.18.1 > >> [4] Biobase_2.16.0 multicore_0.1-7 rtracklayer_1.16.3 > >> [7] GenomicRanges_1.8.9 IRanges_1.14.4 BiocGenerics_0.2.0 > >> [10] reshape2_1.2.1 ggplot2_0.9.1 BiocInstaller_1.4.7 > >> > >> loaded via a namespace (and not attached): > >> [1] BSgenome_1.24.0 Biostrings_2.24.1 DBI_0.2-5 > MASS_7.3-19 > >> [5] RColorBrewer_1.0-5 RCurl_1.91-1 RSQLite_0.11.1 > Rsamtools_1.8.5 > >> [9] XML_3.9-4 biomaRt_2.12.0 bitops_1.0-4.1 > colorspace_1.1-1 > >> [13] compiler_2.15.1 dichromat_1.2-4 digest_0.5.2 > grid_2.15.1 > >> [17] labeling_0.1 lattice_0.20-6 memoise_0.1 > munsell_0.3 > >> [21] plyr_1.7.1 proto_0.3-9.2 scales_0.2.1 > stats4_2.15.1 > >> [25] stringr_0.6.1 tools_2.15.1 zlibbioc_1.2.0 > >> > >> Thanks in advance, > >> > >> Chris > >> > >> [[alternative HTML version deleted]] > >> > >> _______________________________________________ > >> Bioconductor mailing list > >> Bioconductor@r-project.org > >> https://stat.ethz.ch/mailman/listinfo/bioconductor > >> Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > > > [[alternative HTML version deleted]]

ADD REPLY • link 12.3 years ago Valerie Obenchain ★ 6.8k

Login before adding your answer.