Ziad Al Bkhetan

Melbourne, Australia. black.orange@gmail.com

(Replace black and orange by the corresponding part of my name without spaces).

I am a researcher in bioinformatics, machine learning and data science. I received my PhD in 2021 from the School of Computing and Information Systems, the University of Melbourne. My research interests involve applying computational methods, including machine learning and statistical modelling, to genomic problems such as genetic associations with phenotypes and the prediction of chromatin interactions. In addition, I have strong software engineering expertise including web/desktop application development, databases and GPU programming.

Interests and Expertise

Machine Learning and Data Science
  • Data cleansing and preparation for analysis including data migration from relational databases to data warehouse.
  • Experience in classification, regression, clustering and association rules.
  • Feature selection and dimensionality reduction.
  • Statistical assessment and multiple test correction.

Bioinformatics and Computational Biology
  • Genetic association with diseases (Genotype/Haplotype).
  • Population based haplotype phasing or estimation.
  • Chromatin Interaction prediction and three-dimensional genomic structure.

Skills

Programming Languages
  • Currently: R and Python
  • I also have developed projects using: C++, C#, Java and php

Database management Systems
  • Mainly: MySQL
  • I also have used: MS SQL Server, and ORACLE

GPU Progaramming
  • CUDA

Software

eQTLHap

eQTLHap is a comprehensive eQTL analysis tool that scans the genome for three kinds of associations:

  1. Associations with single SNP similar to standard eQTL analysis.
  2. Associations with a block of SNPs encoded by their phased haplotypes.
  3. Associations with a block of SNPs encoded by their genotypes.
eQTLHap is implemented in R and it depends on matrices operations to calculate correlation coefficients. It adapts the ultra-fast Matrix eQTL to deal with blocks and to maintain high speed.

July 2020

consHap

consHap is a majority voting approach to construct consensus haplotype estimator from multiple phased haplotypes.

July 2020

3DEpiLoop

3DEpiLoop is a tool for chromatin Interactions prediction.

February 2018