We will review the revolution brought about by low cost next generation sequencing in a wide array of diagnostic and industrial applications with a special emphasis on computational requirements and big data challenges.
Next generation sequencing ((NGS) has fundamentally changed modern biological research. It is, in fact, an excellent example of how gradual improvements on a powerful initial idea, Sanger’s original dideoxynucleotide sequencing, can lead to such levels of quantitative increase in data production that fundamentally changes a given research field.
Virtually any nucleic acid related research question can be investigated in a comprehensive, high resolution fashion free from experimental confounding factors such as nucleic acid cross hybridization. This has produced a deluge of data on the scale of hundreds of Terabytes even for a single research laboratory. This review will survey both the various application domains of next generation sequencing and their associated computational and analytical challenges.
It was recognized early on that next generation
sequencing will allow querying both the genome (DNA)
and the transcriptome (RNA) on a wide range of
resolution. The exact sequence of nucleotides (e.g. single
nucleotide polymorphisms, single nucleotide variations)
and the overall architecture of the entire genome can be
determined in a single experiment (one run of whole
genome sequencing) [
A wide array of starting materials can be used for next generation sequencing. Any form of nucleic acid (DNA or RNA), from any sources (from inside the cell, from cell free biological fluids or ancient fragmented DNA) can be sequenced and quantified. Nucleic acids can be preselected as in e.g. whole exome sequencing (exon capture) or by other capture mechanisms such as specific protein beacons in ChipSeq analysis. The variations are virtually unlimited and novel approaches are constantly being added to the toolbox of biological research. This universality has led to an enormous variety of application domains. 3 Application domains of next generation sequencing
Since DNA is universal, next generation sequencing
can be used to detect and investigate any life form from
viruses to humans. This is readily exploited in the various
forms of sequencing based microbial diagnostics and it
has also led to a new research field, metagenomics [
Genome wide association studies are a powerful method to identify germline variants associated with increased disease risk.
Remarkably, germline DNA of the fetus can be
efficiently detected in maternal blood, leading to the
powerful tool of prenatal testing [
Cancer is a genetic disease. Accumulating mutations
at various levels of the germline genome lead to
malignant transformation. It is therefore, obvious, that
one of the main targets of next generation sequencing is
cancer diagnostics. Both germline and somatic mutations
are readily identified by NGS. A great number of
oncogenic mutations, many of them targetable by
therapy, have thus been identified [
Recently, sequence analysis of liquid biopsies, essentially cell free DNA obtained from various bodily fluids, emerged as a minimally invasive tool to obtain vital information about the presence of cancer in a patient individual before or during therapy. 4 Computational and analytical challenges of next generation sequencing data
The uniform nature of the biochemical reactions and
ingenuity of technical development has led to an
unexpected situation. The price drop/throughput increase
of next generation sequencing has significantly outpaced
Moore’s law over the past decade. Therefore, next
generation sequencing has become more of a
computational rather than a biochemical problem. The
speed of data accumulation is so fast that data storage and
data analysis is becoming a more and more challenging
problem in modern sequencing based projects. Whole
genome sequencing on a single cancer sample can easily
take up hundreds of GBs of data storage. Therefore, a
single study, such as the one analyzing the whole genome
of 560 breast cancer cases [
Management of vast amounts of data is only one, mainly technical aspect of the challenges at hand.
While next generation sequencing based genomics easily qualifies as one of the main areas of big data science, in many aspects it is also markedly different from those. While in, e.g., financial data the individual variables are connected by poorly understood causative factors and in physics the entire data space is regulated by well defined, homogenous laws of physics, in biology, genomics the situation lies somewhere in between. Variables, such as genes, proteins etc. are connected by the principles of physical chemistry, but the actual parameters of those significantly vary across the various pairs of biological entities. This fact places the analysis of biological systems in the realm of robust, complex systems for which the analytical principles are poorly understood. Therefore, in order to effectively analyze the massive amounts of genomic information one needs to “front-load” the computational analysis with as much biological knowledge as possible. We will present several strategies along those lines. In particular, we will discuss how genomics, next generation sequencing based whole genome analysis helps us to understand DNA repair pathway aberrations, and their diagnostic and therapeutic implications in cancer. We will also discuss how genomics is exploited to understand the main principles of therapeutic immune responses against cancer and how genomics, machine learning and high throughput screening are combined in an interdisciplinary environment to design effective vaccines against cancer. 5 The industrial impact of next generation sequencing
In order to satisfy the need for NGS based diagnostics a whole industry has developed during the past decade. Conferences such as the 2017 Next Generation Dx Summit, (http://www.nextgenerationdx.com) provide an excellent overview of the major trends and players.