Bioinformatic Pipeline Development for Detection of Novel Haemoglobin Variants from Next-Generation Sequencing Data in Nigerian Clinical Laboratories: An Original Computational Haematology Contribution

This dissertation develops and validates an original bioinformatic pipeline for detecting novel and rare haemoglobin variants from next-generation sequencing (NGS) data in Nigerian clinical laboratory settings, addressing the absence of Africa-specific variant detection tools calibrated for the complex haemoglobin variant landscape of the Nigerian population. Rare and novel haemoglobin variants remain diagnostically invisible in standard electrophoresis-based screening, and even HPLC misclassifies structural haemoglobin variants not present in its calibration library, creating a diagnostic gap with clinical consequences for haemolytic anaemia workup, neonatal screening, and genetic counselling. The dissertation combines wet laboratory variant generation and dry laboratory pipeline development across four phases. Targeted amplicon sequencing of the alpha and beta globin gene clusters was optimised for clinical sample DNA quality ranges using Illumina MiSeq. A reference library of 140 known Nigerian haemoglobin variants confirmed by Sanger sequencing and mass spectrometry was constructed. The novel pipeline (HaemVarNig) integrates GATK haplotype calling, custom variant annotation against the globin gene database, and a machine learning pathogenicity classifier trained on Nigerian variant features. HaemVarNig was validated on 300 clinical samples with known genotypes and 80 samples with previously uncharacterised variants. Sensitivity was 99.1 percent and specificity 97.4 percent for known variants. Seventeen novel variants were identified in the uncharacterised samples, seven of which were functionally validated as pathogenic by haematological correlation. HaemVarNig is provided as an open-source tool for African clinical genomics laboratories. Keywords: haemoglobin variants, next-generation sequencing, bioinformatics, clinical laboratory, Nigeria.