Bioinformatics: A Practical Handbook of Next Generation Sequencing and Its Applications

Book Preface

Several years ago, when the first draft of the human genome was being completed, I decided to focus my efforts on the study of pathogen genomes. Armed with a background in software engineering, one of the first things that preoccupied me was a problem that loomed on the horizon and had little to do with the fascinating biology that was emerging from the study of genomes. It was already clear that, in order to study genetic variations, their effects on phenotype, and their epidemiological dynamics, it would be necessary to collect massive amounts of data, far more than most of us could actually handle. The question was not so much whether storage or processing capabilities would be sufficient — Moore’s Law had accustomed us to rapid growth in computing power, and I was confident these technical challenges could be met. The critical question was whether the people who would be analysing these data would have sufficient know-how and resources to handle these large quantities of data, and extract the knowledge they needed. To be sure, the same problem was faced by companies that needed to build search engines, hotel booking systems, web-based ratings software, and all the services based on what we now call “big data”. But genomics looked like a problem that could not be tackled by computer scientists alone. Biologists had to be empowered to handle scary amounts of data.

Those issues were evident even before whole-genome sequencing was revolutionized by the next-generation sequencing (NGS) technologies introduced by companies such as Solexa (now Illumina). Today, the MalariaGEN genomic epidemiology project on which I work (malariagen. net/projects/p-falciparum-community-project) comprises the genomes of Plasmodium parasites from almost ten thousand clinical samples, each backed by several gigabytes of short-read sequencing data — far more data than I would have predicted a few years ago. And yet, the knowledge gap has not been properly filled: if anything, it has become increasingly harder for life scientists and clinicians to effectively process such massive quantities of data, and many projects rely on collaborations with informatics specialists who often have limited expertise of the biological domain. In the light of these difficulties, I give full credit to Lloyd Low and Martti Tammi for making a significant contribution towards filling the gap. What they have produced is a very practical guide, part reference and part tutorial, that will be appreciated by many life scientists for its direct and straightforward approach. Crucially, the content of this book is based on years of teaching experience, and “fine-tuned” by keeping in mind the difficulties routinely faced by those learning how to deal with NGS data. It contains a useful toolkit of techniques and practices using some of the most popular tools in use, such as BWA, samtools and so on. The material covered in this book will support a broad range of applications: the final chapter suggests some possibilities, but clearly each reader will have to tackle challenges unique to their own areas of study, and this work will serve as a base on which to build further techniques. Commendably, it promotes the definition of a well-organized analytical workflow, and gives prominence to the quality aspects of genomics work — hugely important and frequently underestimated.

Conducting a GWAS — or constructing a phylogeny — without first properly evaluating what data to rely upon and what to discard will invariably lead to useless or false results. It is therefore essential to instil high standards of quality into the mind of students and anyone undertaking genomic analyses.

I wish all readers all the best in their endeavours in this complex field, which I hope they will find rich in rewards.

Olivo Miotto