Demuxlet with QCed VCF file

demuxlet_all_qc <- getAllDemuxletResults('190110_demuxlet_new_genotypes/demuxlet_out/demuxlet_plink_bed-updated')

  sample  SNG DBL AMB df_snp$N.SNP
1      1 3074 359   0     89.41159
2      2 5220 311   0     78.10541
3      3 5761 289   0     90.71223
4      4 4058 432   2     51.57124
5      5 6659 675  31     40.57733
[1] "Total number of SNG, DBL and AMB:"

  AMB   DBL   SNG 
   33  2066 24772 
[1] "Total average N.SNP: 67.6666294518254"

Past versions of unnamed-chunk-4-1.png

Past versions of unnamed-chunk-5-1.png

Past versions of unnamed-chunk-5-2.png

Past versions of unnamed-chunk-5-3.png

Past versions of unnamed-chunk-5-4.png

Past versions of unnamed-chunk-5-5.png

Past versions of unnamed-chunk-5-6.png

p

Past versions of fig1-1.png

Demuxlet with QCed VCF, exons only

demuxlet_all_qc_exons <- getAllDemuxletResults('190110_demuxlet_new_genotypes/demuxlet_out/demuxlet_plink_bed-updated.exon_only.recode')

  sample  SNG DBL AMB df_snp$N.SNP
1      1 3056 377   0     68.64433
2      2 5153 376   2     56.22528
3      3 5721 329   0     66.24860
4      4 3957 518  17     35.26870
5      5 6460 749 156     26.79063
[1] "Total number of SNG, DBL and AMB:"

  AMB   DBL   SNG 
  175  2349 24347 
[1] "Total average N.SNP: 48.4977112872614"

Past versions of unnamed-chunk-6-1.png

Past versions of unnamed-chunk-7-1.png

Past versions of unnamed-chunk-7-2.png

Past versions of unnamed-chunk-7-3.png

Past versions of unnamed-chunk-7-4.png

Past versions of unnamed-chunk-7-5.png

Past versions of unnamed-chunk-7-6.png

p

Past versions of fig2-1.png

Demuxlet with QCed VCF + imputed SNPs

demuxlet_all_qc_imputed <- getAllDemuxletResults('190110_demuxlet_new_genotypes/demuxlet_out/demuxlet_chr1_22_combined.qc_r2_maf.recode')

  sample  SNG  DBL AMB df_snp$N.SNP
1      1 2948  485   0    146.00524
2      2 5065  465   1    147.65811
3      3 5593  457   0    165.16992
4      4 4015  477   0    379.63268
5      5 5846 1519   0     86.84942
[1] "Total number of SNG, DBL and AMB:"

  AMB   DBL   SNG 
    1  3403 23467 
[1] "Total average N.SNP: 173.501804919802"

Past versions of unnamed-chunk-8-1.png

Past versions of unnamed-chunk-9-1.png

Past versions of unnamed-chunk-9-2.png

Past versions of unnamed-chunk-9-3.png

Past versions of unnamed-chunk-9-4.png

Past versions of unnamed-chunk-9-5.png

Past versions of unnamed-chunk-9-6.png

p

Past versions of fig3-1.png

Demuxlet with QCed VCF + imputed SNPs, exons only

demuxlet_all_qc_exons_imputed <- getAllDemuxletResults('190110_demuxlet_new_genotypes/demuxlet_out/demuxlet_chr1_22_combined.qc_r2_maf_exon.recode')

  sample  SNG  DBL AMB df_snp$N.SNP
1      1 2890  543   0     75.05884
2      2 5211  320   0    223.39487
3      3 5474  576   0     76.20165
4      4 3714  778   0    154.23753
5      5 6139 1226   0    104.61779
[1] "Total number of SNG, DBL and AMB:"

  DBL   SNG 
 3443 23428 
[1] "Total average N.SNP: 127.186892932902"

Past versions of unnamed-chunk-10-1.png

Past versions of unnamed-chunk-11-1.png

Past versions of unnamed-chunk-11-2.png

Past versions of unnamed-chunk-11-3.png

Past versions of unnamed-chunk-11-4.png

p

Past versions of fig4-1.png

demuxlet_list <- list()
for (i in 1:5){
   demuxlet <- getDemuxletForSample(i, '190110_demuxlet_new_genotypes/demuxlet_out/demuxlet_plink_bed-updated.exon_only.recode')
   demuxlet_list[[i]] <- demuxlet
}
demuxlet_all <- do.call(rbind, unname(demuxlet_list))
 
demuxlet_all$label <- as.character(demuxlet_all$BEST)
demuxlet_all$label[startsWith(demuxlet_all$label, 'DBL')] <- 'DBL'
demuxlet_all$label[startsWith(demuxlet_all$label, 'AMB')] <- 'AMB'
rownames(demuxlet_all) <- demuxlet_all$correct_barcode
 
data <- AddMetaData(data, as.vector(demuxlet_all['label']))
TSNEPlot(data, group.by='label', pt.size=0.1)

Past versions of unnamed-chunk-12-1.png

data@meta.data$label[is.na(data@meta.data$label)] <- "AMB"

depot <- unlist(lapply(data@meta.data$label, function(x){
  if (x == 'SNG-13a_13a'){
    return('Subq')
  } else if (x == 'SNG-1AF_1AF'){
    return('Peri')
  } else if (x == 'SNG-44B_44B'){
    return('Visce')
  } else if (x == 'SNG-BAT14_BAT14'){
    return('Supra')
  } else {
    return('DBL')
  }
}))

data@meta.data['depot'] <- depot
DimPlot(data, reduction='tsne', group.by='depot', pt.size=0.1)

Filter out doublets:

data <- SubsetData(data, cells.use=rownames(data@meta.data)[data@meta.data$depot != 'DBL'])

DimPlot(data, reduction='tsne', group.by='depot', pt.size=0.1)

Past versions of unnamed-chunk-15-1.png

type <- unlist(lapply(as.vector(data@meta.data$depot), function(x){
  if (x == 'Subq' || x == 'Visce'){
    return('white')
  } else {
    return('brown')
  }
}))
data@meta.data['type'] <- type
DimPlot(data, reduction='tsne', group.by='type', pt.size=0.1)

Add depot labels and save

#data@meta.data['depot'] <- substr(data@meta.data$sample_name, 1, nchar(data@meta.data$sample_name)-2)
#saveRDS(data, '../../10x-adipocyte-analysis/output/10x-180831')

Summary of results

demuxlet_all_qc['vcf'] <- 'qc'
demuxlet_all_qc_exons['vcf'] <- 'qc_exons'
demuxlet_all_qc_imputed['vcf'] <- 'qc_imputed'
demuxlet_all_qc_exons_imputed['vcf'] <- 'qc_exons_imputed'
demuxlet_all <- rbind(demuxlet_all_qc, demuxlet_all_qc_exons, demuxlet_all_qc_exons_imputed, demuxlet_all_qc_imputed)

test <- aggregate(demuxlet_all['sng_dbl_amb'], by=list(demuxlet_all$vcf, demuxlet_all$sng_dbl_amb), FUN=length)

df_aggregated <- aggregate(demuxlet_all[c('SNG', 'DBL', 'AMB')], by=list(VCF=demuxlet_all$vcf, timepoint=demuxlet_all$sample), FUN=sum)

df_sng_dbl_amb_vcf <- aggregate(demuxlet_all[c('SNG', 'DBL', 'AMB')], by=list(VCF=demuxlet_all$vcf), FUN=sum)

Number of SNPs, SNG, DBL, AMB

df_snps_vcf <- aggregate(demuxlet_all['N.SNP'], by=list(VCF=demuxlet_all$vcf), FUN=mean)
df_sng_dbl_amb_vcf['N.SNP'] <- df_snps_vcf$N.SNP
as.data.frame(df_sng_dbl_amb_vcf)

               VCF   SNG  DBL AMB     N.SNP
1               qc 24772 2066  33  67.66663
2         qc_exons 24347 2349 175  48.49771
3 qc_exons_imputed 23428 3443   0 127.18689
4       qc_imputed 23467 3403   1 173.50180

df_snps_vcf_timepoint <- aggregate(demuxlet_all['N.SNP'], by=list(VCF=demuxlet_all$vcf, timepoint=demuxlet_all$sample), FUN=mean)

ggplot(df_snps_vcf_timepoint, aes(x=VCF, y=N.SNP, fill=factor(timepoint))) + 
  geom_bar(stat='identity', position='dodge') +
  labs(title='Number of SNPs', fill='timepoint', x='VCF file', y='') +
  theme_gray() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) 

Past versions of unnamed-chunk-20-1.png

Number of singlets, doublets and ambiguous per VCF file.

ggplot(df_aggregated, aes(x=VCF, y=SNG, fill=factor(timepoint))) + 
  geom_bar(stat='identity', position='dodge') +
  labs(title='Number of singlets', fill='timepoint', x='VCF file', y='') +
  theme_gray() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Past versions of unnamed-chunk-21-1.png

ggplot(df_aggregated, aes(x=VCF, y=DBL, fill=factor(timepoint))) + 
  geom_bar(stat='identity', position='dodge') +
  labs(title='Number of doublets', fill='timepoint', x='VCF file', y='') +
  theme_gray() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Past versions of unnamed-chunk-22-1.png

ggplot(df_aggregated, aes(x=VCF, y=AMB, fill=factor(timepoint))) + 
  geom_bar(stat='identity', position='dodge') +
  labs(title='Number of ambiguous', fill='timepoint', x='VCF file', y='') +
  theme_gray() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Past versions of unnamed-chunk-23-1.png

Percentage doublets

print('QCed')

[1] "QCed"

sum(demuxlet_all_qc$DBL) / length(demuxlet_all_qc$DBL)

[1] 0.07688586

print('QCed, exons only')

[1] "QCed, exons only"

sum(demuxlet_all_qc_exons$DBL) / length(demuxlet_all_qc_exons$DBL)

[1] 0.08741766

print('QCed + imputed')

[1] "QCed + imputed"

sum(demuxlet_all_qc_imputed$DBL) / length(demuxlet_all_qc_imputed$DBL)

[1] 0.1266421

print('QCed + imputed, exons only')

[1] "QCed + imputed, exons only"

sum(demuxlet_all_qc_exons_imputed$DBL) / length(demuxlet_all_qc_exons_imputed$DBL)

[1] 0.1281307

#

Session information

sessionInfo()

R version 3.5.3 (2019-03-11)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Storage

Matrix products: default
BLAS/LAPACK: /usr/lib64/libopenblas-r0.3.3.so

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] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] tidyr_0.8.3   dplyr_0.8.0.1 Seurat_2.3.4  Matrix_1.2-17 cowplot_0.9.4
[6] ggplot2_3.1.0

loaded via a namespace (and not attached):
  [1] Rtsne_0.15          colorspace_1.4-1    class_7.3-15       
  [4] modeltools_0.2-22   ggridges_0.5.1      mclust_5.4.3       
  [7] rprojroot_1.3-2     htmlTable_1.13.1    base64enc_0.1-3    
 [10] fs_1.2.7            rstudioapi_0.10     proxy_0.4-23       
 [13] npsurv_0.4-0        flexmix_2.3-15      bit64_0.9-7        
 [16] mvtnorm_1.0-10      codetools_0.2-16    splines_3.5.3      
 [19] R.methodsS3_1.7.1   lsei_1.2-0          robustbase_0.93-4  
 [22] knitr_1.22          jsonlite_1.6        Formula_1.2-3      
 [25] workflowr_1.2.0     ica_1.0-2           cluster_2.0.7-1    
 [28] kernlab_0.9-27      png_0.1-7           R.oo_1.22.0        
 [31] compiler_3.5.3      httr_1.4.0          backports_1.1.3    
 [34] assertthat_0.2.1    lazyeval_0.2.2      lars_1.2           
 [37] acepack_1.4.1       htmltools_0.3.6     tools_3.5.3        
 [40] igraph_1.2.4        gtable_0.3.0        glue_1.3.1         
 [43] reshape2_1.4.3      RANN_2.6.1          Rcpp_1.0.1         
 [46] trimcluster_0.1-2.1 gdata_2.18.0        ape_5.3            
 [49] nlme_3.1-137        iterators_1.0.10    fpc_2.1-11.1       
 [52] gbRd_0.4-11         lmtest_0.9-36       xfun_0.5           
 [55] stringr_1.4.0       irlba_2.3.3         gtools_3.8.1       
 [58] DEoptimR_1.0-8      MASS_7.3-51.1       zoo_1.8-5          
 [61] scales_1.0.0        doSNOW_1.0.16       parallel_3.5.3     
 [64] RColorBrewer_1.1-2  yaml_2.2.0          reticulate_1.11.1  
 [67] pbapply_1.4-0       gridExtra_2.3       rpart_4.1-13       
 [70] segmented_0.5-3.0   latticeExtra_0.6-28 stringi_1.4.3      
 [73] foreach_1.4.4       checkmate_1.9.1     caTools_1.17.1.2   
 [76] bibtex_0.4.2        Rdpack_0.10-1       SDMTools_1.1-221   
 [79] rlang_0.3.2         pkgconfig_2.0.2     dtw_1.20-1         
 [82] prabclus_2.2-7      bitops_1.0-6        evaluate_0.13      
 [85] lattice_0.20-38     ROCR_1.0-7          purrr_0.3.2        
 [88] labeling_0.3        htmlwidgets_1.3     bit_1.1-14         
 [91] tidyselect_0.2.5    plyr_1.8.4          magrittr_1.5       
 [94] R6_2.4.0            snow_0.4-3          gplots_3.0.1.1     
 [97] Hmisc_4.2-0         pillar_1.3.1        whisker_0.3-2      
[100] foreign_0.8-71      withr_2.1.2         fitdistrplus_1.0-14
[103] mixtools_1.1.0      survival_2.43-3     nnet_7.3-12        
[106] tsne_0.1-3          tibble_2.1.1        crayon_1.3.4       
[109] hdf5r_1.1.1         KernSmooth_2.23-15  rmarkdown_1.12     
[112] grid_3.5.3          data.table_1.12.0   git2r_0.25.2       
[115] metap_1.1           digest_0.6.18       diptest_0.75-7     
[118] R.utils_2.8.0       stats4_3.5.3        munsell_0.5.0