Genetic Tests

Introduction

I specialise in developing new genetic tests for clinical laboratories. This page lists the tests I have worked on so far in my career, starting with the most recent.

Developing and delivering genetic tests is a joint effort between clinical scientists, bioinformaticians, technologists, technicians, data-scientists, clinicians and administrative staff. All of the projects listed below were collaborations with many wonderful colleagues without whom these projects would not have succeeded.

Chromosome-arm-level Copy Number Variant Detection in Solid Cancers

In this project we validated detection of copy number variants which impact entire chromosome arms, which are increasingly used in the diagnosis and stratification of brain tumours. This included the recurrent 1p19q co-deletion and simultaneous gain of chromosome 7 with loss of chromosome 10 (+7/-10).

Project Overview

Outcome: diagnostic service

Timing: May 2025-Jan 2026

Location: Manchester Genomic Laboratory Hub

Technique: QIAseq next generation sequencing

Code: R and Quarto

Skills I learned

  • How to create interactive plots in R and python using the plotly package
  • How to combine multiplex ligation-dependent probe amplification (MLPA) data and next generation sequencing (NGS) data in R

Deletions, LOH and Ploidy Detection in Solid Cancers

This project was the third section of a broader project to validate copy number variant detection for solid cancers at the North West GLH. It was the most technically complex validation I’ve done so far, involving over 150 samples, 47 different genes and 4 different laboratory methods, with investigation of gene deletions, loss of heterozygosity (LOH) and different chromosomal ploidy states.

Project Overview

Outcome: diagnostic service

Timing: Nov 2024-April 2025

Location: Manchester Genomic Laboratory Hub

Technique: QIAseq next generation sequencing

Code: R and Quarto

Skills I learned

  • How to create my first R package: foldchanger
  • How to perform automated code testing with testthat
  • How to perform validation comparisons across multiple techniques and data layouts, including whole genome sequencing, droplet digital PCR, short tandem repeat PCR and NGS

Amplifications of 9 Oncogenes in Diverse Cancers

This project built on my previous validation of ERBB2 amplifications, adding a further 8 genes: ALK, BRAF, EGFR, MET, MDM2 MYC, MYCN and SMO. These genes are important for prognosis and treatment planning in many different cancer types.

Project Overview

Outcome: diagnostic service

Timing: Apr 2024-Nov 2024

Location: Manchester Genomic Laboratory Hub

Technique: QIAseq next generation sequencing

Code: R

Skills I learned

  • How to build a shiny app for quality monitoring
  • How to render full documentation and analysis with R and Quarto into a Word template
  • How to manipulate whole genome sequencing HTML results using rvest

ERBB2 Amplifications in Solid Cancers

This project was to detect amplifications of the ERBB2 (HER2) gene in solid tumour samples, focusing on lung, colorectal and brain cancers. ERBB2 is an important gene in many cancer types, and is targeted by several medications including trastuzumab (Herceptin).

Project Overview

Outcome: diagnostic service

Timing: Dec 2023-Apr 2024

Location: Manchester Genomic Laboratory Hub

Technique: QIAseq next generation sequencing

Code: R

Skills I learned

  • How to create CNV plots with embedded gene, exon and primer locations
  • How to combine R and Quarto for analysis
  • How to perform limit of detection experiments with reference materials

Genomic Instability Testing for Ovarian Cancer

In this project I validated a new bioinformatics pipeline for detection of genomic instability in ovarian cancer, which was developed by the company SeqOne Genomics. Accurate detection of genomic instability is necessary for identifying optimum-responders to the anti-cancer drug olaparib (Lynparza®). This workflow is now in diagnostic service, and the results are used to plan patient treatment.

Project Overview

Outcome: diagnostic service

Timing: Sept 2023-Dec 2023

Location: Manchester Genomic Laboratory Hub

Technique: shallow whole genome sequencing (sWGS)

Code: R

Skills I learned

  • How to analyse sWGS data
  • How to query an SQL database with dbplyr using an ODBC connection
  • How to parse information from PDFs with pdftools using regular expressions and raw strings

Spinocerebellar Ataxia type 36

Spinocerebellar Ataxia type 36 (SCA36) is a very rare neurological disorder caused by a hexanucleotide repeat expansion in the NOP56 gene, with fewer than 200 patients reported in the literature. I optimised a repeat-primed PCR protocol and validated it on a cohort of samples tested via whole genome sequencing with the ExpansionHunter algorithm. This diagnostic service is, to my knowledge, the first in the world for this condition.

Project Overview

Outcome: diagnostic service

Timing: Nov 2022-July 2023

Location: University College London Hospital Neurogenetics Department

Technique: repeat-primed PCR and flanking PCR

Code: R

Skills I learned

  • How to design repeat-primed PCR primers
  • How to optimise repeat-primed PCRs
  • How to integrate a literature review into a validation document

RFC1-Disorder Testing

RFC1-disorder is a neurological disorder caused by repeat expansions in the RFC1 gene, and was first reported in 2019 by Dr Andrea Cortese’s research group at the UCLH Institute for Neurology. For this project I developed a diagnostic service for RFC1 expansions by modifying the available research protocol, and successfully added RFC1 testing to the NHS National Genomic Test Directory.

Project Overview

Outcome: diagnostic service

Timing: Jan 2022-July 2023

Location: University College London Hospital Neurogenetics Department

Technique: multiple repeat-primed PCRs and flanking PCR

Code: R

Skills I learned

  • How to fully optimise a flanking PCR to detect large amplicons
  • How to apply to the National Genomic Test Directory for new test indication
  • How to integrate multiple PCRs into a single workflow for increased efficiency and cost-effectiveness

Mosaicism Detection for Rare Skin Disorders

The aim of this project was to confirm mosaic sequence variants for rare skin disorders using droplet digital PCR (ddPCR). The main challenges were that the variants were sometimes present in 1% of the sample or lower, the DNA quantities from skin biopsies were limited, and many of the variants were only found in a single patient. The solution was to use R to collate all the ddPCR data for multiple assays and establish uniform thresholds for background normal variation. The technique was then brought into diagnostic service, and over 100 reports were issued in the first 6 months.

Project Overview

Outcome: diagnostic service

Timing: Oct 2021-March 2022

Location: North London Genomic Laboratory Hub

Technique: ddPCR

Code: R

Skills I learned

  • How to set data-driven analysis thresholds for ddPCR
  • How to use RMarkdown to generate validation plots and tables
  • How to automatically audit clinical services using RMarkdown scripts

Non-Invasive Prenatal Diagnosis for Sickle Cell Disease and Rare Genetic Conditions

In this project I developed a ddPCR test for sickle cell disease which could diagnose the condition prenatally using only a maternal blood sample. The project was then expanded to developing bespoke ddPCR assays for rare genetic conditions with different inheritance patterns. The correct prediction rate was optimised to 98% in comparison with invasive testing, but the presence of 2 incorrect results prevented the implementation of the method as a diagnostic service, and we published our findings in the journal Prenatal Diagnosis.

Project Overview

Outcome: publication

Timing: September 2019-January 2022

Location: North London Genomic Laboratory Hub

Technique: droplet digital PCR

Code: R and JAGS

Skills I learned

  • How to write an original research paper
  • How to write code in R using tidyverse packages
  • How to optimise ddPCR assays

Myotonic Dystrophy Type 1

This was my first validation project, which I began as a trainee clinical scientist. Myotonic dystrophy type 1 is caused by a triplet repeat expansion in the DMPK gene, and testing for this condition is a core requirement for genetics laboratories in England. I was asked to validate the test at the North London Genomic Laboratory Hub as part of my clinical scientist training. The validation successfully passed external evaluation by the United Kingdom Accreditation Service (UKAS) and went into diagnostic service.

Project Overview

Outcome: diagnostic service

Timing: July 2018-Jan 2020

Location: North London Genomic Laboratory Hub

Technique: repeat-primed PCR

Skills I learned

  • How to write a validation document to the standards of ISO:15189
  • How to calculate mobility shifts in capillary electrophoresis using GeneMarker
  • How to write standard operating procedures for wet lab work and analysis