Human health in the tropics

Tropical and infectious diseases represent a significant health burden to people living in the tropics. However, increasingly affordable and portable high-throughput molecular technologies are already being trialled to address issues in tropical and infectious disease. For example, an increasing number of clinical metagenomics programs are being trialled for pathogen identification both in hospitals and in the field. Such programs typically yield whole genome sequences which offer improvements over existing assay-based methods in that they are unbiased and can detect novel pathogens as well as identify AMR genes and be used in phylodynamic studies. Vaccine development programs are also increasingly incorporating high-throughput data to better understand the underlying genetic mechanisms which is important in the tropics with only 8 of the 20 WHO defined tropical diseases offering commercially licensed vaccines [11]. Molecular epidemiology is another growing application with advances in portable and rapid long-read sequencing in particular changing how we detect and track infectious diseases. The potential to develop near real-time diagnostics using these technologies represents a huge opportunity to development in-house capabilities in infectious disease detection and treatment.

In an ever-increasing number of applications, high-throughput sequencing data is being used to alleviate this burden, largely in the areas of pathogen identification/characterisation, molecular epidemiology, vaccine development, host-pathogen interactions and precision medicine programs (Table 1).

Table 1 Applications of Omics Technologies to address Health in the tropics i)

Pathogen identification and characterisation

The ability to detect causative pathogens driving human disease is an essential precursor to diagnosis and effective treatment. Pathogens causing disease are more common in the tropics and challenging to detect, especially in regional and remote health care settings. Current pathogen identification techniques can be time consuming and imprecise, and critically are unable to detect novel pathogens, rendering them unsuitable for emerging pathogen surveillance that is required in any outbreak scenario. These limitations mean patients are often treated prior to diagnosis with long-duration and broad-spectrum antimicrobials, which contributes to the spread of drug-resistant microbes. Metagenomic next-generation sequencing (mNGS) offers an emerging solution to this problem as it has been successfully used in a research environment to diagnose complex clinical cases where traditional diagnostic tests have failed using both short-read [12] and long-read sequencing [28]. An advantage of this approach is ability to completely reconstruct the genome with high-resolution sequence data enabling strain identification [29], virulence gene detection [19] and phylodynamic studies [17]. While unlikely to replace isolate sequencing (due to host and commensal bacteria contamination), mNGS offers an exciting opportunity to simultaneously detect viruses, bacteria, parasites and fungi in a single test in situations where isolates cannot be obtained, or where health care infrastructure is limited.

Pathogen discovery is also an important tool in predicting future outbreaks and pandemics with the majority of viruses still unknown. The number of known viruses has grown by orders of magnitude since the advent of metagenomic-based high-throughput sequencing. These discoveries are occurring across the tree of life with a recent study identifying 1445 new RNA viruses in invertebrates, including some that are sufficiently divergent to comprise new families [30]. These studies are not limited to understudied organisms however with 140,000 viruses just characterised across ~ 28,000 human microbiome samples, over half of which have never been seen before [31]. Knowledge of the wider viral landscape is critical for monitoring potential outbreak scenarios, with COVID-19 demonstrating how devastating such events can be. The problem is especially acute in the tropics where the virome of most tropical species is largely uncharacterised [13].

ii)

molecular epidemiology

New sequencing technologies, especially portable and rapid long-read sequencing is driving important improvements to the way molecular epidemiology is applied in the tropics. Recent outbreaks of Ebola and COVID-19 have demonstrated how near real-time sequencing of pathogens is critical to help better understand both the evolution and transmission of viruses [32]. Recent work by Faust et al. [33] describes their experience utilising long-read sequencing to perform molecular field research on schistosomiasis, trypanosomiasis and rabies for the benefit of the local communities in Uganda. With long reads we can improve diagnostics, better understand disease transmission dynamics and ideally provide feedback to endemic communities regarding clear actionable timelines. Whole-genome sequencing (WGS) is also being used to inform about antimicrobial resistance in near real-time. A recent program in the Philippines incorporated WGS within the established Antimicrobial Resistance Surveillance Program to better understand resistance mechanisms and local transmission patterns [18]. Their work linking resistance phenotypes to WGS data revealed the mixing of genetic strains and subsequent AMR mechanisms which identified the AMR vehicles responsible for driving the expansion of increasing carbapenem resistance rates within the country.

iii)

vaccine development

The development of effective vaccines to combat the wide range of tropical and infectious diseases is increasingly relying on sequence information to inform design strategies. With only 8 of 41 NTDs offering commercially licensed vaccines, work is needed to incorporate sequence information to enable rational systems-level vaccine design [34]. A perfect vaccine offers ongoing protection from a pathogen by eliciting both innate and adaptive immunity, however we typically have little understanding of how the host and pathogen interact at a systems level. High-throughput sequencing data offer us insight into how both pathogen and host respond to vaccination and infection [35]. These datasets provide insight into the underlying mechanisms driving immunological memory, protection offered by the vaccine and the efficiency of both antigens and adjuvants. Numerous programs are successfully employing this approach for diseases such as tuberculosis [36, 37], malaria [38] and dengue fever [39].

iv)

host-pathogen interactions

The interaction between host and pathogen is recognised as critical to how both populations and individuals respond to infectious disease. Genome sequencing of host and pathogen offers insight into the genomic determinants of adaptive processes driving parasite virulence and host resistance (e.g. malaria [40]). Using this approach, typically genetic variants are identified across both host and pathogen (within populations or even individuals) and the association of these variants with the observed phenotype used to reduce the genomic search space by identifying the driving genetic determinants [41]. Another important application of host-pathogen interactions is detecting proof of horizontal gene transfer, a process whereby an organism acquires genetic material from distantly related species in order to gain new functional capabilities [42]. While common in prokaryotes [43], there are examples of eukaryotes acquiring new functionality from parasites [44]. While much remains to be discovered regarding host-pathogen interactions in the tropics, detailed genomic sequence data offer unprecedented insight into this process.

xxii)

Precision medicine

Precision medicine programs are increasingly being rolled out worldwide using a combination of targeted-, exome- and whole-genome sequencing to identify clinically actionable variants in patients [45, 46]. While challenges exist [47], such programs have led to increasingly accurate diagnosis and management of genetic disease. Critical to successful precision medicine programs is the ability to incorporate patient genetic information to develop custom treatment options. There are several challenges that need to be addressed to roll out such programs in tropical regions, largely the need for supporting sequencing infrastructure / analysis capability and the lack of detailed population-specific variant information. The lack of infrastructure requires investments in both lab and analytical capability within health care settings, with most precision medicine programs currently using lab-intensive short-read sequencing capabilities.

Long-read sequencing shows great promise in this space as it requires less infrastructure and can be used in remote locations. Long-read sequencing has now been used to identify disease causing variants [48] and recently researchers at Stanford were able to successfully diagnose a patient in a matter of hours in a critical care setting [49]. Thus, the future of long-read sequencing for disease diagnosis is promising. Another challenge for the tropical population is the lack of ethnically matched allele frequencies within the variant repositories for screening, with this information being critical in prioritising candidate disease causing variants [50]. Without using ethnic matching, uncatalogued population-specific genetic variants may be incorrectly characterized as novel largely due to under-representation of the patient’s ethnic group within the repositories. Despite efforts to increase the representation of peoples from tropical regions, more work is needed to incorporate genetic information from all groups including Indigenous populations [27]. Sequencing efforts such as the Centre for Indigenous Genomics in Australia will begin to address this disparity.

Biodiversity

The biodiversity of the tropics is unparalleled in terms of biomass and numbers of species; however, this diversity is likely underestimated as many tropical species remain undescribed. Not only do we need to know and understand our tropical biodiversity in order to protect it for its own intrinsic value but cataloguing and ultimately preserving tropical biodiversity is critical to maintaining clean water, air and soil and to regulate the climate, recycle nutrients and provide food. Tropical forests currently absorb 15% of the anthropogenic carbon emissions and are critically important for mitigating the effects of climate change [51]. Maintaining biodiversity also represents a huge opportunity with a large ecological reservoir of natural products including antibiotics, many of which have been used as traditional medicines for millennia. Despite traditional usage however, in most cases the active product remains unknown, and more work is needed to identify the active molecules and their mode of action. Antibiotics are of particular interest to address the global challenge of antimicrobial resistance (AMR), which is projected to kill 10 million people annually by 2050 [52, 53].

The tropics are also home to the world’s most diverse marine (coral reefs) and terrestrial (rainforest) ecosystems. In addition to the relational and cultural values that are associated with biodiversity [54], reefs and rainforests help sustain both tropical and non-tropical human populations through the provision of direct ecological services such as food, freshwater and erosion mitigation, and through global-scale effects such as carbon sequestration, climate modulation and as habitat for long-ranging migratory species [55]. Biodiversity is directly linked to ecosystem health and robustness which in turn are the major determinants of an ecosystem’s capacity to support human populations [56, 57]. Current threats to tropical biodiversity include land clearing, over exploitation, climate change, ocean acidification, habitat fragmentation and invasive species [58,59,60], however modern sequencing technologies provide us with new tools to document, monitor, analyse and advise upon the best strategies to mitigate the worst effects of these threats (Table 2).

Table 2 Tropical Biodiversity Techniques and applications i)

Biodiversity Discovery

With much of tropical biodiversity understudied and under described, large scale efforts to describe as many species as efficiently as possible are key. Metagenomics and DNA metabarcoding are two technologies that can find unique genomes and unique species from environmental samples. These techniques are particularly useful because traditional taxonomy is unable to detect cryptic species, which are particularly common in the tropics. The International Barcode of Life Consortium (iBOL) has renewed its efforts to describe all of Earth’s biodiversity, with the 7-year BIOSCAN project (2019–2026) [61] promising to analyse more than 10 million specimens.

ii)

Conservation genomics

The advent and refinement of low cost long-read sequencing technologies has emboldened researchers to adopt increasingly ambitious goals regarding the application of bioinformatic technologies to the fields of biodiversity research and conservation. The past decade has seen a proliferation of large consortia that aim to provide high-quality genome assemblies for selected cohorts of species with the goal of providing critical resources for species’ conservation and management. Most major initiatives currently underway can be categorised as either region specific (e.g. 1,000 Chilean Genomes, Darwin Tree of Life [77], European Reference Genome Atlas [78]) or taxon specific (e.g. 10,000 Plant Genomes [64], Global Invertebrate Genomics Alliance [79], Genome 10 K [80]).

At a higher level, in 2018 the Earth BioGenome Project (EBP) was proposed with the explicit aim of sequencing all eukaryotic life on earth [62, 63]. This well-resourced initiative adopted an umbrella-style organisational model through which they aim to facilitate sequencing projects undertaken by member organisations. At the time of writing, over forty consortia are listed as partner members to the EBP, however only four regional initiatives (Africa BioGenome Project [5], AusARG, Oz Mammals Genomics Initiative [81], Taiwan BioGenome Project) cover tropical areas. While most of the taxon-specific initiatives include tropical representatives, in practice tropical species remain underrepresented in the rapidly growing list of sequenced genomes. The current enthusiasm for large-scale genome sequencing projects coupled with the relative lack of sequenced tropical species represents a huge opportunity for future tropical biodiversity-focused sequencing efforts.

iii)

threatened species management

The value of genome sequencing to conservation efforts ranges from the assessment of the genetic health of species and the planning of breeding programs to the establishment of naturally occurring variant databases for the implementation of genetic rescue plans of fragmented populations [78, 82, 83]. Recent conservation efforts focusing on the Tasmanian Devil [84] and the kākāpō [85] highlight the value of leveraging a high quality reference genome for planning, implementing and managing conservation strategies.

For species as yet without genomic resources, the use of SNP sequencing and eDNA has been able to advance conservation measures for many tropical species. For example, SNP sequencing has been used to understand the effective population size and genetic connectivity between breeding sites for the critically-endangered Kuranda Treefrog [69] and thereby inform management priorities for this species. Environmental DNA (eDNA) is being used to investigate the true range of threatened species such as the critically-endangered Armoured Mistfrog in north-east Australia [70] and the endangered big-headed turtle Platysternon megacephalum in Hong Kong [86]. With current projections of tropical habitat fragmentation and species loss set to continue in tropical regions [87,88,89], genome-centric management strategies will likely garner increasing adoption over the coming century.

iv)

invasive species management

Invasive species also pose severe risks to tropical biodiversity. In 1991 the Golden Mussel (Limnoperna fortunei (Dunker, 1857)), native to East Asia, was introduced to the Rio Del Plata in Argentina through the expulsion of ship ballast water [90]. This invasive freshwater bivalve mollusc, which was first sequenced in 2018 [91], spread rapidly throughout Argentina, Uruguay, Paraguay and Brazil from where it now risks entering the Amazon river basin [92]. Traditional methods of monitoring of the golden mussel across such an expansive territory would be prohibitively resource intensive, however a recent effort to utilise eDNA to detect traces of Limnoperna DNA from water samples has proven to be an effective and cost-efficient means of monitoring [72].

The use of eDNA for indirect detection of terrestrial tropical species is now well-established [93] and is currently being implemented for monitoring the expansion of invasive species such as the cane toad Rhinella marina in Australia [94, 95] and the detection of Yellow Crazy Ants (Anoplolepis gracilipes) invading the tropical rainforests and streams of north-east Australia [73] through sampling freshwater streams.

xxii)

Drug discovery

Tropical marine and rainforest species are exceptionally rich sources of bioactive compounds. Historically, most new drug discoveries came through the targeting of natural compounds present in traditional medicinal plant species [96]. This ‘ethnobotanical’ approach to drug discovery remains a leading source of new bioactive compounds with up to 50% of all natural product-derived drugs having plant-based origins [96].

Bioprospecting for new drug candidates has long involved the targeting of plant, fungi and microbe species but starting in the 1950s, attention turned to marine invertebrates following the discovery of spongothymidine and spongouridine from the tropical marine sponge Tethya crypta. Modern derivatives of these compounds are now powerful antiviral and anticancer drugs [97, 98]. Today sponges and cnidarians account for up to 80% of all invertebrate-based drug discoveries with the remaining 20% coming from a wide range of other phyla [99]. This ‘ecological’ approach to drug discovery, in combination with the previously described ‘ethnobotanical’ approach, constitute the two main components of the ‘biorational’ drug discovery strategy [34].

Although these candidate-based approaches are undoubtedly effective, many of the steps involved in traditional drug discovery and validation pipelines are laborious and costly. While laboratory-based validation of potential drug candidates cannot be replaced, in silico modelling and data mining offer the potential of substantial efficiency gains by narrowing the focus of wet-lab assays and experiments [100, 101]. Following the isolation of a candidate compound, 3D molecular modelling, determination of target affinity, prediction of off-target effects and reverse engineering of biosynthetic pathways are all steps that today benefit from the implementation of bioinformatic pipelines to supplement lab-based experiments and assays. By coupling indigenous knowledge and domain-specific ecological knowledge with modern high-throughput biotechnological and computational approaches, the resulting efficiency gains are beginning to have a positive impact on the enormous costs associated with bringing drugs to market [102,103,104].

Food production

The tropical climate is ideal for agriculture (crops, livestock, aquaculture) with its high temperature all year round, high rainfall and sun exposure [97]. Unfortunately, these benefits are often counter balanced in many tropical areas due to poor soil fertility, inadequate water supply for irrigation, lack of mechanization and the high prevalence of diseases [7]. Historically, food production in the tropics has been characterised by lower yields and higher variability than in temperate regions despite favourable growing conditions [