Variant Summary

Summary results for variants called by MINTIE on Leucegene normals.

summary_by_sample <- normals_results %>%
                        group_by(sample) %>%
                        summarise(variants = length(unique(variant_id)))
summary_by_class <- collate_vartypes(normals_results) %>%
                        group_by(class) %>%
                        summarise(variants = length(unique(variant_id))) %>%
                        mutate(fraction = variants / sum(variants))
print(summary_by_sample)

# A tibble: 23 x 2
   sample    variants
   <chr>        <int>
 1 SRX372044      181
 2 SRX372045      137
 3 SRX372046      213
 4 SRX372047      639
 5 SRX372048     1119
 6 SRX372049      143
 7 SRX372050      497
 8 SRX372051      449
 9 SRX372052      235
10 SRX372053       75
# … with 13 more rows

print(summary_by_class)

# A tibble: 4 x 3
  class                          variants fraction
  <chr>                             <int>    <dbl>
1 Fusion                               69   0.0117
2 Novel splice variant               1992   0.338 
3 Transcribed structural variant     1234   0.210 
4 Unknown                            2593   0.440 

Variant Genes

MINTIE paper Figure 4 showing the number of variant genes called across the Leucegene normal samples.

var_genes <- normals_results$overlapping_genes %>% 
                str_split("\\||:")

repeat_rows <- rep(1:nrow(normals_results), sapply(var_genes, length))
results_by_gene <- data.table(normals_results[repeat_rows,])
results_by_gene$gene <- unlist(var_genes)

results_summary <- results_by_gene[, length(unique(gene)), by = c("sample", "cell_type")]
results_summary <- results_summary %>% arrange(cell_type, V1) %>% data.table()
results_summary$sample <- factor(results_summary$sample,
                                 levels = results_summary$sample)

print("Summary of variant genes called across samples:")

[1] "Summary of variant genes called across samples:"

summary(results_summary$V1)

   Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
   81.0   130.5   155.0   186.7   218.0   496.0 

# reorder cell types by total
results_totals <- results_summary[, sum(V1), by = c("cell_type")] %>%
                    arrange(desc(V1))
results_summary$cell_type <- factor(results_summary$cell_type,
                                    levels = results_totals$cell_type)

ggplot(results_summary, aes(cell_type, V1, group=sample)) + 
    geom_bar(position = position_dodge2(width = 0.9, preserve = "single"), stat = "identity") +
    theme_bw() +
    xlab("") +
    ylab("Genes with variants")

Past versions of plot_variant_genes-1.png

Library Size and Variant Number Correlation

Perform correlation calculation on the library size and number of variant genes found per sample.

Leucegene Gene Expression notebook must be run before this chunk to generate the expression counts matrix.

# load counts data, calculate library sizes and add to results summary
counts <- fread(here("output/Leucegene_gene_counts.tsv"))
libsizes <- apply(counts, 2, sum) %>% data.frame()
colnames(libsizes) <- "libsize"
libsizes$sample <- factor(rownames(libsizes),
                          levels = results_summary$sample)
results_summary <- left_join(results_summary, libsizes, by ="sample", "sample")

lib_var_cor <- cor(results_summary$libsize, results_summary$V1, method = "spearman")
print(paste("Spearman correlation between library size and variant genes called:", lib_var_cor))

[1] "Spearman correlation between library size and variant genes called: 0.140845074722306"

ggplot(results_summary, aes(libsize, V1, colour = cell_type)) +
    geom_point() +
    theme_bw() +
    ylab("Genes with variants")

Past versions of libsize_var_cor-1.png

Protein Coding Genes

Proportion of protein coding genes observed in the MINTIE results.

# load CHESS gene reference containing gene types
chess_genes <- read.delim(gzfile(here("data/ref/chess2.2.genes.gz")))

# join gene info with results and summarise by gene type
gene_count <- left_join(results_by_gene, chess_genes, by = c("gene" = "Gene_Name")) %>%
                group_by(Gene_Type) %>%
                summarise(n_genes = length(unique(gene))) %>%
                data.table()

n_protein_coding <- gene_count[gene_count$Gene_Type == "protein_coding", "n_genes"]
print(paste("proportion of protein coding genes =", n_protein_coding / sum(gene_count$n_genes)))

[1] "proportion of protein coding genes = 0.81547619047619"

Controls Comparison

MINTIE Supplementary Figure 3 showing variant genes called in Leucegene Total White Blood Cell samples with different cell types as control groups.

# get TWBC results
controls_comp <- normals_results[normals_results$cell_type == "Total white blood cells",]
controls_comp$controls <- "twbc"
controls_comp$cell_type <- NULL

# load comparisons against all other controls
controls_test_dir <- here("data/leucegene/normals_controls_test_results")
celltypes <- list.files(controls_test_dir)
for (celltype in celltypes) {
    tmp <- str_c(controls_test_dir, celltype, sep = "/") %>%
            list.files(., full.names = TRUE) %>%
            lapply(., read.delim) %>%
            rbindlist() %>%
            filter(logFC > 5) %>%
            mutate(controls = celltype)
    controls_comp <- rbind(controls_comp, tmp)
}

# extract variant genes and make summary
var_genes <- controls_comp$overlapping_genes %>%
                str_split("\\||:")

repeat_rows <- rep(1:nrow(controls_comp), sapply(var_genes, length))
results_by_gene <- data.table(controls_comp[repeat_rows,])
results_by_gene$gene <- unlist(var_genes)

results_summary <- results_by_gene[, length(unique(gene)), by = c("sample", "controls")]
results_summary <- results_summary %>% arrange(sample, V1) %>% data.table()
results_summary$sample <- factor(results_summary$sample,
                                 levels = unique(results_summary$sample))

# reorder by totals across different controls
results_totals <- results_summary[, sum(V1), by = c("controls")] %>%
                    arrange(V1)
results_summary$controls <- factor(results_summary$controls,
                                    levels = results_totals$controls)

print("Total variant genes called using different controls:")

[1] "Total variant genes called using different controls:"

print(results_summary)

       sample controls  V1
 1: SRX372044     mono 187
 2: SRX372044     twbc 193
 3: SRX372044     gran 279
 4: SRX372044       tc 329
 5: SRX372044       bc 365
 6: SRX372045     gran 130
 7: SRX372045     mono 138
 8: SRX372045     twbc 156
 9: SRX372045       bc 345
10: SRX372045       tc 378
11: SRX372046     mono 192
12: SRX372046     twbc 233
13: SRX372046     gran 247
14: SRX372046       tc 419
15: SRX372046       bc 443

ggplot(results_summary, aes(sample, V1, fill=controls)) +
    geom_bar(position=position_dodge2(width=0.9, preserve="single"), stat="identity") +
    theme_bw() +
    xlab("") +
    ylab("Genes with variants") +
    scale_fill_brewer(palette = "RdYlBu",
                      labels =  c("mono" = "Monocytes",
                                  "twbc" = "Total white blood cells",
                                  "gran" = "Granulocytes",
                                  "tc" = "T-Cells",
                                  "bc" = "B-Cells"))

Past versions of controls_comparison-1.png

Session information

sessionInfo()

R version 3.5.1 (2018-07-02)
Platform: x86_64-apple-darwin15.6.0 (64-bit)
Running under: macOS  10.14.6

Matrix products: default
BLAS: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRblas.0.dylib
LAPACK: /Library/Frameworks/R.framework/Versions/3.5/Resources/lib/libRlapack.dylib

locale:
[1] en_AU.UTF-8/en_AU.UTF-8/en_AU.UTF-8/C/en_AU.UTF-8/en_AU.UTF-8

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

other attached packages:
[1] edgeR_3.24.3      limma_3.38.3      ggplot2_3.1.0     stringr_1.4.0    
[5] here_0.1          dplyr_0.8.1       data.table_1.12.0

loaded via a namespace (and not attached):
 [1] Rcpp_1.0.1         RColorBrewer_1.1-2 plyr_1.8.4        
 [4] compiler_3.5.1     pillar_1.3.1       later_1.0.0       
 [7] git2r_0.26.1       workflowr_1.6.1    tools_3.5.1       
[10] digest_0.6.18      lattice_0.20-38    evaluate_0.13     
[13] tibble_2.1.1       gtable_0.3.0       pkgconfig_2.0.2   
[16] rlang_0.4.2        cli_1.1.0          yaml_2.2.0        
[19] xfun_0.5           withr_2.1.2        knitr_1.22        
[22] fs_1.2.7           locfit_1.5-9.1     rprojroot_1.3-2   
[25] grid_3.5.1         tidyselect_0.2.5   glue_1.3.1        
[28] R6_2.4.0           fansi_0.4.0        rmarkdown_1.12    
[31] purrr_0.3.2        magrittr_1.5       whisker_0.3-2     
[34] backports_1.1.3    scales_1.0.0       promises_1.1.0    
[37] htmltools_0.3.6    assertthat_0.2.1   colorspace_1.4-1  
[40] httpuv_1.5.2       labeling_0.3       utf8_1.1.4        
[43] stringi_1.4.3      lazyeval_0.2.2     munsell_0.5.0     
[46] crayon_1.3.4