GenomeAnalysis

The Genome Analysis Pipeline enables accurate variant detection and evaluation using powerful bioinformatics tools like GATK, Samtools, FastQC, with reference genomes hg19 or GRCh38.

1. Data Preprocessing

Data preprocessing is a crucial step to ensure high-quality input data before variant analysis.

• Quality Control (QC): Use FastQC to assess the quality of sequencing reads.
• Trimming & Filtering: Remove low-quality bases and adapter sequences using tools like Trimmomatic or Cutadapt.
• Alignment (Mapping): Align reads to the reference genome (hg19 or GRCh38) using BWA-MEM or Bowtie2, generating BAM/SAM files.
• Sorting & Indexing: Sort and index BAM files using Samtools or Picard.
• Mark Duplicates: Remove duplicate reads with Picard MarkDuplicates to reduce bias.
• Base Quality Score Recalibration (BQSR): Adjust base quality scores using GATK BaseRecalibrator to minimize sequencing errors.

2. Germline Short Variant Discovery (GATK)

The goal is to identify SNPs (Single Nucleotide Polymorphisms) and Indels (Insertions & Deletions) from human genome data.

• HaplotypeCaller: Use GATK HaplotypeCaller for variant calling.
• CombineGVCFs: Merge multiple GVCF files from different samples into a single file.
• GenotypeGVCFs: Convert GVCF into VCF containing identified variants.
• Variant Filtration: Apply quality filters such as DP (Depth), QD (Quality by Depth), MQ (Mapping Quality) using GATK VariantFiltration.

3. Callset Refinement

After variant calling, refinement is needed to improve accuracy.

• Variant Recalibration (VQSR): Use GATK VariantRecalibrator to refine variant quality using a machine learning model.
• Hard Filtering: If no training data is available, apply hard filtering criteria (e.g., QD < 2.0, FS > 60.0, MQ < 40).
• Normalize Variants: Use vt normalize to standardize variant representation.

4. Callset Evaluation

Evaluating the variant callset ensures data quality and pipeline correctness.

• Concordance Analysis: Compare against gold-standard variant datasets like dbSNP, 1000 Genomes, gnomAD.
• Functional Impact Analysis: Assess variant effects using SnpEff or ANNOVAR.
• Variant Density Plots: Check the distribution of variants across the genome using bcftools stats and plot-vcfstats.

5. Annotate Variants

Variant annotation provides biological and clinical insights.

• VEP (Variant Effect Predictor): Predicts the impact of variants on genes and proteins.
• dbSNP & ClinVar Annotation: Cross-reference with clinical databases to identify disease-related variants.
• Pathogenicity Prediction: Use PolyPhen and SIFT to assess variant pathogenicity.

6. Somatic Short Variant Discovery

Identifying somatic variants, especially in cancer data, by comparing tumor vs normal samples.

• Mutect2 (GATK): Detect somatic variants from paired tumor-normal data.
• FilterMutectCalls: Remove false positives to improve variant accuracy.
• MSI & TMB Analysis: Analyze Microsatellite Instability (MSI) and Tumor Mutational Burden (TMB) to characterize tumor profiles.

TOOLS and Data samples

Link: https://drive.google.com/drive/folders/1DQJ3x0bPmIOXtw440rLcE-fYX5JZZY7B?usp=sharing