HAF-pipe

HAF-pipe estimates haplotype-inferred allele frequencies from pooled next-generation sequencing (pool-seq) data to enable allele frequency tracking in evolve-and-resequence (E+R) experiments.

Key Features:

• Low-Coverage Sequencing Efficiency: Enables accurate allele frequency estimation from ultra-low coverage (<5x) pooled sequencing compared to traditional high-coverage (>100x) approaches.
• Haplotype Inference: Infers known founder haplotypes within small genomic windows to improve accuracy for bi-allelic SNPs, especially in populations founded from sequenced homozygous strains.
• Robustness to Missing Data and Recombination: Maintains accuracy under moderate missing data and across up to 50 generations of recombination, validated with experimentally-pooled and simulated samples including Drosophila melanogaster.
• Predictive Modeling: Uses a simple linear model derived from simulations to predict accuracy of haplotype-derived allele frequencies across different organisms and experimental designs.

Scientific Applications:

• Evolve-and-Resequence (E+R) experiments: Tracking allele frequency dynamics over time to detect adaptive alleles and study evolutionary responses in experimental populations.
• Experimental planning and validation: Informing replication and sequencing depth choices for E+R studies and validating performance using simulated pooled-seq data.

Methodology:

Processing of raw sequencing data is implemented with a combination of bash and R scripts, and simulation capabilities are provided by HAFpipe-sim.run_forqs.sh and HAFpipe-sim.simulate_poolseq.sh to simulate recombination and pooled sequencing data from sequenced founder panels.