Key Deliverables & Statistical Frameworks
Whole-genome and exome variant discovery, population genetics, structural variation, copy-number profiling, and somatic cancer genomics.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 001 — Manhattan Plot
Use Case Application: Identifies genome-wide significant susceptibility loci associated with complex traits or disease phenotypes across chromosomes.
Required Input Data: GWAS summary statistics table containing Chromosome, Base-Pair Position, SNP rsID, and association p-value across millions of variants.
Statistical & Algorithmic Method: Generalized Linear Mixed Models (GLMM) or logistic/linear regression; negative log10 p-value transformation against Bonferroni threshold (-log10(5e-8) = 7.30).
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 002 — Miami Plot
Use Case Application: Compares genetic association signals symmetrically between two cohorts, sexes, ancestries, or correlated clinical phenotypes.
Required Input Data: Two aligned GWAS summary statistic datasets sharing identical genomic coordinates (Chromosome, Position, p-values for Cohort A and Cohort B).
Statistical & Algorithmic Method: Mirrored -log10(p) transformation (+log10 for upper track, -log10 for lower track) using stratified mixed-model association test statistics.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 003 — GWAS Quantile-Quantile (Q-Q) Plot
Use Case Application: Diagnoses genomic inflation, cryptic population stratification, or polygenic signal enrichment across genome-wide test statistics.
Required Input Data: Vector of observed genome-wide association p-values or chi-square test statistics from cohort regression.
Statistical & Algorithmic Method: Order-statistic quantile comparison against theoretical Uniform(0,1) null distribution; genomic inflation factor lambda_GC = median(chi2_obs) / 0.4549.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 004 — Circos Inter-Chromosomal Translocation Plot
Use Case Application: Visualizes structural rearrangements, gene fusions, copy-number tracks, and inter-chromosomal translocations in cancer genomes.
Required Input Data: BEDPE structural variant breakpoint calls (chr1, pos1, chr2, pos2, SV type) paired with chromosome karyotype ideogram coordinates.
Statistical & Algorithmic Method: Discordant paired-end read clustering and split-read breakpoint assembly (e.g., Manta/Delly) mapped via quadratic Bezier chords in polar coordinates.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 005 — Linear Genome Browser Coverage Track
Use Case Application: Inspects localized sequencing read depth, exon-intron coverage uniformity, and candidate variant pileups across target loci.
Required Input Data: Coordinate-sorted BAM/CRAM alignment files or BigWig normalized read-depth tracks paired with GTF/GFF3 gene annotations.
Statistical & Algorithmic Method: Per-base pileup depth calculation normalized by Counts Per Million (CPM) or Reads Per Kilobase Million (RPKM) across sliding genomic bins.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 006 — Sashimi Alternative Splicing Plot
Use Case Application: Quantifies exon skipping, intron retention, and differential isoform usage between experimental conditions.
Required Input Data: RNA-seq BAM alignments and splice-junction BED/tab files containing exon read densities and junction-spanning split read counts.
Statistical & Algorithmic Method: Percent Spliced In (PSI) estimation via Dirichlet-multinomial or Beta-binomial hierarchical modeling (rMATS / LeafCutter) with splice-arc counts.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 007 — Whole-Genome Synteny Dot Plot
Use Case Application: Detects large-scale chromosomal inversions, duplications, translocations, and conserved syntenic blocks between species or assemblies.
Required Input Data: Pairwise whole-genome alignment coordinates (PAF, MUMmer delta, or BLAST tabular start/end positions on Query vs. Reference).
Statistical & Algorithmic Method: Maximal Unique Match (MUM) chaining and dynamic programming collinearity scoring across homologous genomic intervals.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 008 — Marey Recombination Rate Map
Use Case Application: Contrasts physical genomic distance (Mb) against genetic linkage distance (cM) to pinpoint recombination hotspots and centromeric suppression.
Required Input Data: Integrated genetic linkage map markers containing physical coordinates (Mb) and meiotic crossover distances (centiMorgans).
Statistical & Algorithmic Method: Monotonic cubic spline or LOESS smoothing of genetic vs. physical position; local recombination rate (cM/Mb) derived via first derivative.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 009 — Chromosome Giemsa Ideogram
Use Case Application: Maps cytogenetic G-banding patterns, centromeres, telomeres, and annotated disease loci onto physical chromosome scaffolds.
Required Input Data: Cytoband annotation table (chromosome, start, end, band name, Giemsa stain intensity) and genomic feature BED coordinates.
Statistical & Algorithmic Method: Proportional linear scaling of base-pair intervals mapped to standardized ISCN cytogenetic staining palettes.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 010 — Copy Number Variation (CNV) Segmentation Plot
Use Case Application: Detects somatic or germline focal amplifications, homozygous deletions, and chromosomal arm-level aneuploidies.
Required Input Data: Tumor-normal paired or panel-normalized log2 read-depth ratios across contiguous genomic bins.
Statistical & Algorithmic Method: Circular Binary Segmentation (CBS) or Hidden Markov Models (HMM) partitioning noisy log2-ratio profiles into piecewise-constant copy states.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 011 — Somatic Mutation OncoPrint
Use Case Application: Summarizes multi-gene somatic alteration landscapes, co-occurrence, and mutual exclusivity across patient cohorts.
Required Input Data: MAF (Mutation Annotation Format) somatic variant calls and discrete GISTIC copy-number matrices across genes and samples.
Statistical & Algorithmic Method: Binary/categorical alteration matrix sorted by gene mutation frequency and mutual exclusivity (Fisher's exact test / CoMEt algorithm).
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 012 — Protein Domain Lollipop Plot
Use Case Application: Pinpoints recurrent somatic hotspot mutations relative to functional protein domains and active catalytic sites.
Required Input Data: Amino acid substitution coordinates (HGVSp), variant functional classifications, and UniProt/Pfam domain boundaries.
Statistical & Algorithmic Method: Positional mutation recurrence counting with binomial or Poisson background mutation rate testing for hotspot enrichment.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 013 — Kataegis Rainfall Plot
Use Case Application: Reveals localized hypermutation clusters (kataegis) driven by APOBEC cytidine deaminase activity in tumor genomes.
Required Input Data: Somatic single-nucleotide variant (SNV) genomic positions and pyrimidine-context base substitution classes.
Statistical & Algorithmic Method: Inter-mutational distance computation (IMD = pos_i - pos_{i-1}) plotted on a log10 scale along chromosomal coordinates.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 014 — Mutational Signature Trinucleotide Profile
Use Case Application: Deconstructs tumor mutational catalogs into underlying exogenous (UV, tobacco) and endogenous (HRD, MMR) mutational processes.
Required Input Data: VCF somatic SNVs stratified into 96 trinucleotide sequence contexts (6 substitution types x 16 flanking 5'/3' bases).
Statistical & Algorithmic Method: Non-Negative Matrix Factorization (NMF) decomposing sample mutation counts V into signature matrix W and exposure matrix H.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 015 — Information-Content Sequence Logo
Use Case Application: Characterizes transcription factor binding motifs, splice donor/acceptor consensus sequences, and conserved protein sites.
Required Input Data: Multiple sequence alignment (MSA) or Position Weight Matrix (PWM) of nucleotide or amino acid frequencies per position.
Statistical & Algorithmic Method: Shannon entropy uncertainty H(l) = -sum(p_b * log2(p_b)); total stack height R_seq = log2(4) - (H(l) + e_n) measured in bits.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 016 — Site / Allele Frequency Spectrum (SFS)
Use Case Application: Infers demographic history, population bottlenecks, purifying selection, and rare variant burden in population cohorts.
Required Input Data: Multi-sample VCF genotype calls across unrelated individuals within a defined population.
Statistical & Algorithmic Method: Derived or minor allele count histogram folded or unfolded across 2N chromosomes, compared against neutral coalescent expectation 1/k.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 017 — Linkage Disequilibrium (LD) r² Matrix
Use Case Application: Delimitates haplotype blocks and identifies tag SNPs in non-random allelic association across fine-mapped loci.
Required Input Data: Phased or unphased diploid genotype matrix (0, 1, 2) for biallelic variants within a target genomic window.
Statistical & Algorithmic Method: Pairwise squared correlation of allele frequencies r² = D² / (p_A * p_a * p_B * p_b), where D = p_AB - p_A * p_B.
[DOMAIN 1: GENOMICS, VARIANT CALLING & COMPARATIVE GENOMICS]
Plot 018 — Tajima's D Sliding-Window Selection Scan
Use Case Application: Detects signatures of balancing selection (D > 0) or positive selective sweeps / population expansion (D < 0).
Required Input Data: Aligned population DNA sequences or multi-sample VCF calls partitioned into overlapping genomic windows.
Statistical & Algorithmic Method: Normalized difference between mean pairwise nucleotide differences (pi) and Watterson's estimator (theta_W = S / a_1), standardized by variance.