From Data to Decisions: Why Genomics and AMR Need Practical Bioinformatics
Genomics has transformed how we study pathogens. Today, sequencing a bacterial genome is faster and cheaper than ever. Yet, despite this progress, one fundamental challenge remains. The real bottleneck is turning raw reads into a decision a lab, clinician, or IPC team can act on fast, reproducibly, and with confidence.
This question sits at the heart of my work, and it’s the reason I started this blog.
The data–decision gap
In many settings, especially across low- and middle-income countries, we now have access to sequencing platforms, trained personnel, and growing datasets. What often lags behind is the translation layer:
-
How do we ensure data quality before analysis?
Which pipelines are reproducible and sustainable (versioned containers, workflow languages like WDL/Nextflow, clear inputs/outputs)?
Which analysis infrastructure/platform will we use (local HPC vs cloud like Terra/AWS/GCP) so workflows are secure, scalable, and auditable?
How do we interpret AMR signals responsibly (coverage/contamination checks, phenotype–genotype concordance, transparent thresholds)?
How do results feed back into clinical, laboratory, IPC, or policy decisions (report formats, turnaround time, dashboards, action triggers)?
Too often, genomic data stops at figures in a manuscript or reports that never inform action.
Why AMR genomics is different
Antimicrobial resistance is not just a research problem—it is a systems problem. Genomic insights must connect with:
-
microbiology laboratories,
-
clinicians,
-
infection prevention teams,
-
surveillance units, and
-
national policy frameworks.
This requires bioinformatics that is not only technically sound but context-aware.
What this blog will focus on
Here, I will share practical insights drawn from real projects, including:
-
Designing and running reproducible genomic workflows
-
Interpreting “messy” data (mixed samples, contamination, low coverage)
-
AMR gene detection and phylogenetic inference in practice
-
Lessons from deploying pipelines across institutions
-
Training approaches that build long-term capacity
Where possible, I’ll emphasize why certain choices matter not just how to run a tool.
A note on accessibility
I aim to write posts that are:
-
technically accurate,
-
grounded in experience,
-
accessible to early-career scientists,
-
and useful to seasoned practitioners.
Some posts will be hands-on. Others will be reflective. All will be driven by one principle and if you remember one thing from this blog, let it be this:
Genomics only matters if it informs better decisions.
What’s coming next
Upcoming posts will cover:
-
How I assess WGS data quality before any downstream analysis
-
What to do when a sample shows signals of being “mixed”
-
Building an end-to-end AMR analysis workflow
-
Reproducibility lessons from real-world bioinformatics projects
If you work in genomics, AMR, or public health or you’re training in these areas, you are welcome. I hope these posts provide practical workflows, interpretation tips, and implementation lessons you can apply immediately.
If you’d like me to cover a specific tool or workflow (e.g., Kraken2, Snippy, AMRFinderPlus, ARIBA; or reproducible workflows with WDL/Cromwell/Terra), leave a comment below or reach out via my website.
You can also find more of my work here:
Website: https://gmboowa.github.io/
Gerald Mboowa, I work at the intersection of genomics, antimicrobial resistance (AMR), and public health implementation across Africa and global partners with a focus on reproducible workflows and practical decision support.
Dr. Gerald, this post perfectly captures the essential challenge of translating genomic data into actionable decisions that can truly impact antimicrobial resistance control. The focus on reproducible workflows, data quality, and context-aware bioinformatics is critical, especially in settings where resources are limited. I also want to acknowledge the important role of the African Society for Laboratory Medicine (ASLM) in strengthening laboratory capacity and quality assurance across Africa, which is fundamental to enabling sustainable genomic surveillance and AMR decision-making. As someone privileged to be supervised by Dr. Gerald Mboowa, I have firsthand experience of his commitment to bridging the data decision gap and fostering practical, reproducible approaches that empower laboratories and public health teams. His leadership continues to inspire and guide many of us working at the intersection of genomics, AMR, and public health implementation. I look forward to following this blog and applying these valuable information in our ongoing efforts to combat antimicrobial resistance effectively.
ReplyDeleteThank you very much, Robert, for such a thoughtful & generous reflection on the post.
DeleteI’m really glad the emphasis on reproducible workflows, data quality, and context-aware bioinformatics resonated with you, especially from the perspective of implementation in resource-limited settings, where these elements truly determine whether genomics informs action or remains academic. You are absolutely right to highlight the role of the African Society for Laboratory Medicine (ASLM); strong laboratory systems, quality assurance, and workforce development are foundational to sustainable genomic surveillance and meaningful AMR decision-making.
I deeply appreciate your kind words and, more importantly, your engagement at the intersection of genomics, AMR & public health implementation. Seeing trainees and colleagues translate these ideas into practical impact is exactly what motivates this work. I look forward to continuing to learn from your contributions and to seeing how you apply and extend these approaches in your ongoing efforts to combat antimicrobial resistance.
Thank you again for reading, engaging, and advancing this important conversation.
— Gerald