Sliding Window Analysis of \DeltaSNP-index

Description

DeltaSNPWindow compute \DeltaSNP-index values across chromosomes using an overlapping sliding-window approach to identify genomic regions exhibiting allele frequency differentiation between experimental conditions. The method follows the QTL-seq framework described by Takagi et al. (2013), which builds upon the SNP-index strategy introduced by Abe et al. (2012). The function computes the mean \DeltaSNP-index within each window to highlight consistent allele frequency differences and optionally applies a Wilcoxon signed-rank test to evaluate whether window-level deviations significantly differ from zero. Multiple testing correction is performed using the Benjamini–Hochberg method. Regions showing strong and statistically supported \DeltaSNP-index signals represent candidate loci enriched for biologically relevant allele frequency shifts.

Usage

DeltaSNPWindow(
  deltasnp_input,
  delta_cols = NULL,
  window_size = 1e6,
  step_size = 1e5,
  run_wilcoxon = TRUE
)

Arguments

Details

The function:

• Generates sliding windows for each chromosome.

• Calculates mean \DeltaSNP-index within each window.

• Optionally performs Wilcoxon signed-rank test to test deviation from zero.

• Adjusts p-values using Benjamini-Hochberg FDR correction.

Value

A data frame containing:

• seqnames – Chromosome name

• start – Window start position

• end – Window end position

• width – Window width

• strand – Genomic strand

• Mean \DeltaSNP-index columns (prefixed with mean_)

• Wilcoxon p-values (if enabled)

• FDR-adjusted p-values (if enabled)

Author(s)

Sayanti Guha Majumdar, Saima Bano, Nandita Banerjee, Rahul Kumar Tiwari, Dipro Sinha, Subham Ghosh, Sanjeev Kumar

References

Abe, A., Kosugi, S., Yoshida, K., Natsume, S., Takagi, H., Kanzaki, H., Matsumura, H., Yoshida, K., Mitsuoka, C., Tamiru, M., Innan, H., Cano, L. M., Kamoun, S., Terauchi, R. (2012). Genome sequencing reveals agronomically important loci in rice using MutMap. Nature Biotechnology, 30(2), 174–178. doi:10.1038/nbt.2095

Takagi, H., Abe, A., Yoshida, K., Kosugi, S., Natsume, S., Mitsui, H., Uemura, A., Utsushi, H., Tamiru, M., Takahashi, R., Goto, K., Terauchi, R. (2013). QTL-seq: rapid mapping of quantitative trait loci in rice by whole genome resequencing of DNA from two bulked populations. Plant Journal, 74(1), 174–183. doi:10.1111/tpj.12105

Benjamini, Y., and Hochberg, Y. (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B, 57(1), 289–300.

Examples

example_file <- system.file(
  "extdata",
  "All_delta_SNPs_Index.tsv",
  package = "DESNP"
)

window_result <- DeltaSNPWindow(
  deltasnp_input = example_file,
  window_size = 5e5,
  step_size = 1e5,
  run_wilcoxon = TRUE
)

head(window_result)

Compute Delta SNP-index

Description

DeltaSNPindex calculates the difference in SNP-index values between two experimental conditions using output from the SNPindex() function. The Delta SNP-index (\DeltaSNP-index) is computed as the subtraction of SNP-index values between groups for each SNP position, representing allele frequency changes between conditions. This approach is based on the QTL-seq strategy described by Takagi et al. (2013). The function also supports automatic group detection and optional threshold-based filtering to identify SNPs showing substantial allele frequency differentiation.

Usage

DeltaSNPindex(
  snpindex_input,
  meta_file = NULL,
  filter_deltaSNP = TRUE,
  deltaSNP_threshold = 0.5,
  filter_direction  = c("both", "greater", "lesser"),
  column_filter     = c("all", "any")
)

Arguments

Details

\DeltaSNP-index is calculated as:

\Delta SNPindex = SNPindex_{condition2} - SNPindex_{condition1}

Modes of operation:

1. Metadata mode

If meta_file is provided:

• If a group has exactly 2 conditions → single \DeltaSNP-index column

• If more than 2 conditions → all pairwise comparisons are generated

2. Auto mode

If meta_file = NULL, the function:

• Automatically detects control columns (control/ctrl/c)

• Computes treatment vs control \DeltaSNP-index

• Falls back to pairwise comparisons if no control is detected

The function returns all computed \DeltaSNP-index values and optionally filtered results.

Value

Returns a list containing:

all_delta_SNPs

Data frame containing all computed \DeltaSNP-index values

filtered_delta_SNPs

Filtered SNPs based on threshold

summary

Summary statistics of the analysis

Author(s)

Sayanti Guha Majumdar, Saima Bano, Nandita Banerjee, Rahul Kumar Tiwari, Dipro Sinha, Subham Ghosh, Sanjeev Kumar

References

Takagi, H., Abe, A., Yoshida, K., Kosugi, S., Natsume, S., Mitsui, H., Uemura, A., Utsushi, H., Tamiru, M., Takahashi, R., Goto, K., Terauchi, R. (2013). QTL-seq: rapid mapping of quantitative trait loci in rice by whole genome resequencing of DNA from two bulked populations. Plant Journal, 74(1), 174–183. doi:10.1111/tpj.12105

Examples

snp_file <- system.file("extdata", "All_SNPs_Index.tsv",
                        package = "DESNP")

meta_file <- system.file("extdata", "meta_columns.tsv",
                         package = "DESNP")

result_deltasnpindex <- DeltaSNPindex(
  snpindex_input = snp_file,
  meta_file = meta_file,
  deltaSNP_threshold = 0,
  filter_direction = "both",
  column_filter = "any"
)

head(result_deltasnpindex$all_delta_SNPs)
head(result_deltasnpindex$filtered_delta_SNPs)
result_deltasnpindex$summary

SNP-index Calculation

Description

Computes SNP-index values from allele depth (AD) fields of the VCF file, performs statistical comparison between experimental conditions, and identifies significantly differentiated SNPs following the SNP-index framework described by Abe et al. (2012). The function supports beta-binomial regression and Fisher's exact test, with automatic test selection based on replicate availability.

Usage

SNPindex(
  vcf_file,
  metadata_file,
  min_depth = 10,
  test_method = c("auto", "beta_binomial", "fisher_exact"),
  PVALUE = 0.05,
  Pvalue_filter = c("all", "any")
)

Arguments

Details

The SNP-index is calculated as:

SNPindex = ALT / (REF + ALT)

SNPs with total depth below min_depth are removed before testing.

Statistical testing strategy:

• If \ge2 biological replicates are detected per condition, beta-binomial regression is applied.

• If replicates are insufficient, Fisher's exact test is used.

• In "auto" mode, the test is selected automatically.

P-values are adjusted for multiple testing using the Benjamini–Hochberg (BH) method and reported as FDR.

Value

Returns a list containing:

all_SNPs

Data frame of all analyzed SNPs including allele counts, SNP-index values, p-values, and FDR.

significant_SNPs

Subset of SNPs passing the PVALUE threshold.

summary

Summary statistics including total SNPs, number of significant SNPs, test used, depth threshold, and PVALUE.

Author(s)

Sayanti Guha Majumdar, Saima Bano, Nandita Banerjee, Rahul Kumar Tiwari, Dipro Sinha, Subham Ghosh, Sanjeev Kumar

References

Abe, A., Kosugi, S., Yoshida, K., Natsume, S., Takagi, H., Kanzaki, H., Matsumura, H., Yoshida, K., Mitsuoka, C., Tamiru, M., Innan, H., Cano, L. M., Kamoun, S., & Terauchi, R. (2012). Genome sequencing reveals agronomically important loci in rice using MutMap. Nature Biotechnology, 30(2), 174–178. doi:10.1038/nbt.2095

Benjamini, Y., and Hochberg, Y. (1995). Controlling the false discovery rate. Journal of the Royal Statistical Society: Series B, 57(1), 289–300.

See Also

read.vcfR, vglm, fisher.test

Examples

vcf_path <- system.file("extdata", "combined_top10_new_smut.vcf",
                        package = "DESNP")

meta_path <- system.file("extdata", "metadata.tsv",
                         package = "DESNP")


result_snpindex <- SNPindex(
  vcf_file = vcf_path,
  metadata_file = meta_path,
  min_depth = 10,
  test_method = "auto",
  PVALUE = 0,
  Pvalue_filter = "any"
)


head(result_snpindex$all_SNPs)
head(result_snpindex$significant_SNPs)
result_snpindex$summary

Genome-wide Manhattan Plot for SNP Index and \DeltaSNP Index

Description

Generates a customizable genome-wide Manhattan plot for SNP index (SNPindex_) or \DeltaSNP index (delta_snp_index_) values. The function automatically detects appropriate SNP index columns or allows the user to specify columns manually. It computes cumulative genomic positions across chromosomes and produces a ggplot object. Faceting is supported when multiple SNP index columns are plotted.

Usage

manhattan_plot(
  input_data,
  plot_cols = NULL,
  plot_title = "Genome-wide Manhattan Plot of SNP Index",
  x_label = "Chromosome",
  y_label = NULL,
  y_limits = c(0, 1),
  y_breaks = NULL,
  x_expand = 0.01,
  point_size = 0.7,
  point_alpha = 0.7,
  hline_value = 0.5,
  hline_linetype = "dashed",
  hline_color = "black",
  title_size = 14,
  title_face = "bold",
  axis_title_size = 12,
  axis_text_size = 8,
  axis_text_angle = 40,
  facet_ncol = 1,
  facet_scales = "free_y"
)

Arguments

Details

Automatic Column Detection

If plot_cols = NULL, the function:

1. Searches for columns matching ^delta_snp_index(_.*)?$

2. If none found, searches for ^SNPindex_

3. Stops if no matching columns are detected.

Cumulative Genome Position

To create a genome-wide Manhattan plot across multiple chromosomes:

• Chromosome lengths are calculated using maximum POS.

• Cumulative offsets are computed.

• SNP positions are shifted accordingly.

This ensures chromosomes appear sequentially along the x-axis.

Faceting

If more than one SNP index column is plotted:

• Separate panels are created using facet_wrap()

• Y-axis scaling can be controlled via facet_scales

Plot Output

The function returns a ggplot object. Use ggsave() if you want to save PNG, PDF, or TIFF files manually.

Value

A ggplot object representing the Manhattan plot.

Author(s)

Sayanti Guha Majumdar, Saima Bano, Nandita Banerjee, Rahul Kumar Tiwari, Dipro Sinha, Subham Ghosh, Sanjeev Kumar

Examples

example_file <- system.file(
  "extdata",
  "All_delta_SNPs_Index.tsv",
  package = "DESNP"
)

manhattan_plot(input_data = example_file)

# Plot specific column
manhattan_plot(
  input_data = example_file,
  plot_cols = c("delta_snp_index_SNPindex_trt_200dai_vs_SNPindex_control_200dai")
)

# Custom y-limits and horizontal line
manhattan_plot(
  input_data = example_file,
  y_limits = c(-1, 1),
  hline_value = 0
)