Friday 4 December 2020

EdwardsLab at #AusEvo2020

If you missed his talk at ABACBS2020, Jack will be presenting today at the Australasian Evolution Society 2020 Conference about The role of gene duplication in the evolution of snake venoms. Two conference presentations in two weeks - not a bad way to prepare for your Honours viva post-submission. Well done, Jack!

Also, Kat Stuart will be presenting her work on invasive starlings in Zoom 2 at 13:00 AEDT. Kat’s talks are always great to listen to:

  • Katarina Stuart: What drives invasion success? Using historical museum samples to examine evolution in an invasive passerine.

Wednesday 25 November 2020

#ABACBS2020: Unsupervised orthologous gene tree enrichment for cost-effective phylogenomic analysis and a test case on waratahs (Telopea spp.)

Stephanie Chen, Maurizio Rossetto, Marlien van der Merwe, Hervé Sauquet, Patricia Lu-Irving, Jia-Yee Yap, William Studley, Greg Bourke, Jason Bragg, Richard J. Edwards


Whole-genome shotgun sequencing is becoming increasingly common in phylogenetic research due to the falling cost of whole genome sequencing compared to traditional methods which target subsets of genomes. However, there are few existing packages for assembling putatively orthologous loci from evolutionarily diverged samples and making alignments for phylogenetic analysis from these data. Additionally, short-read Illumina sequencing data are highly accurate but at low coverages, it can be difficult to draw out meaningful phylogenomic inferences, especially for non-model organisms for which there is no reference genome available.

We have developed a scalable method of rapidly generating species trees from short-read data without the need for a reference genome. The workflow involves (1) de novo genome assembly with ABySS at a range of k values (2) extracting the most complete BUSCO (Benchmarking Universal Single-Copy Orthologs) genes from each set of assemblies with the BUSCO Compiler and Comparison tool (BUSCOMP) (3) generating gene trees, and (4) constructing a species tree.

The workflow has been applied to a whole genome shotgun sequencing waratah (Telopea spp.) dataset of five species, comprising of two samples from each of the seven lineages; there are three lineages of T. speciosissima (New South Wales waratah) – coastal, upland, and southern. We have also generated a reference genome for T. speciosissima, and examine the robustness of the workflow by comparison to a reference-based approach. It is anticipated that the workflow will maximise the recovery of informative data from genomic datasets for reproducible phylogenomic studies and be especially useful for non-model organisms.

#ABACBS2020: Whole transcripts in genome assembly, annotation, and assessment: the draft genome assembly of the globally invasive common starling, Sturnus vulgaris

Katarina Stuart, Yuanyuan Cheng, Lee Rollins & Richard J. Edwards


Native to the Palearctic, the common starling (Sturnus vulgaris) is a near-globally invasive passerine that has now colonised every continent barring Antarctica. Ecological interest in the species is two-fold – they are considered a conservation risk and crop pest within the invasive ranges, while recent decades have brought with them a worrying decline in starling numbers within historical native ranges. Despite the global interest in this species, there are still fundamental knowledge gaps in our understanding of the genetics and population differences of this species across their native and invasive range. We present the Australian S. vulgaris draft genome and transcriptome to be used as a reference for further investigation into evolutionary characterisation of this ecologically significant species. An initial 10x Genomics linked-read assembly was scaffolded and gap-filled with low coverage nanopore sequencing, complemented by PacBio Isoseq full-length transcript data. Isoseq data was incorporated into assembly scaffolding, annotation, and assembly assessment to inform workflow decisions. We produced a draft assembly with a scaffold N50 size of 72.5 Mb, and assess this alongside a North American S. vulgaris draft genome, previously assembled from Illumina data. Lastly, we use these different reference genomes, alongside a non-scaffolded version of the Australian S. vulgaris genome to assess how choice of reference genome affects common population genetic downstream analysis using a global whole genome resequencing data set.

Tuesday 24 November 2020

#ABACBS2020: The role of gene duplication in the evolution of snake venoms

Jack Clarke, Vicki Thomson & Richard Edwards


Snakes are one of the most venomous animals on the planet, using their venom for defence and the capturing of prey. Snake venoms have evolved independently of other venoms in other vertebrates, and there is considerable variation between species in their proteomic composition. One of the primary mechanisms through which snake venoms are thought to evolve is the duplication, recruitment and specialisation of proteins from other tissues. In some cases, this evolution is known to involve the tandem duplication of genes resulting in chromosomal clusters of venom genes in some gene families. We have recently sequenced and assembled the genomes of two highly venomous Australian snakes: Notechis scutatus (mainland tiger snake) and Pseudonaja textilis (eastern brown snake). In conjunction with publicly available proteomes from 10 other venomous snakes and 2 non-venomous snakes, these genomes provide an excellent opportunity to examine the role that duplication and neofunctionalisation has played in snake venom evolution.

We have analysed 43 protein families known to play a role in snake venom and examined their pattern of duplication in snakes, compared to high quality reference genomes of other reptiles and non-venomous vertebrates. We find evidence for extensive duplications across some of these families, but no clear enrichment for duplication in the evolution of venom specifically. Instead, we identify a trend where numerous duplications specific to venomous snakes occur in proteins that seem predisposed to evolve by duplication and specialisation, even in non-venomous vertebrates. A subset of high-quality snake genomes was then used to further explore the nature of duplications. While tandem gene duplication is evident in some larger families, it remains absent in many.

The snake venom metalloproteinase (SVMP) family provides an excellent case study, with multiple duplication events throughout its evolutionary history in vertebrates. Part of the broader ADAM (“a disintegrin and metalloproteinase”) family of single-pass transmembrane and secreted zinc proteases, SVMP appears to have expanded by independent tandem duplications in different snake lineages. We also identify a second ADAM subfamily, ADAM20, with an abundance of venomous snake-specific duplications. Ongoing work in exploring the possible role of ADAM20 proteins in snake venoms and the role that genome assembly quality has played in our ability to robustly detect the presence or absence of gene duplication events.

Monday 2 November 2020

Antarctic desert soil bacteria exhibit high novel natural product potential, evaluated through long-read genome sequencing and comparative genomics

The third of our collaborative “controlled bacterial metagenome” de novo whole genome assembly projects was published in Environmental Microbiology in November. This was a fun collaboration with the Ferrari lab at UNSW trying to maximise bang for buck to sequence some complete bacterial genomes using PacBio sequencing to identify biosynthetic gene clusters. As with a previous paper, we used pooled genomic DNA sequencing and were able to assemble complete genomes (and plasmids) of the 13/17 species that had sufficient depth of coverage. Coolest of all (if you excuse the pun), these were bugs from an Antarctic expedition! Head over to the Ferrari lab website to find out more about their research.

Benaud N, Edwards RJ, Amos TG, D’Agostino PM, Gutiérrez-Cháveza C, Montgomery K, Nicetic I & Ferrari BC (2020). Antarctic desert soil bacteria exhibit high novel natural product potential, evaluated through long-read genome sequencing and comparative genomics. Environmental Microbiology.


Actinobacteria and Proteobacteria are important producers of bioactive natural products (NP), and these phyla dominate in the arid soils of Antarctica, where metabolic adaptations influence survival under harsh conditions. Biosynthetic gene clusters (BGCs) which encode NPs, are typically long and repetitious high G + C regions difficult to sequence with short‐read technologies. We sequenced 17 Antarctic soil bacteria from multi‐genome libraries, employing the long‐read PacBio platform, to optimize capture of BGCs and to facilitate a comprehensive analysis of their NP capacity. We report 13 complete bacterial genomes of high quality and contiguity, representing 10 different cold‐adapted genera including novel species. Antarctic BGCs exhibited low similarity to known compound BGCs (av. 31%), with an abundance of terpene, non‐ribosomal peptide and polyketide‐encoding clusters. Comparative genome analysis was used to map BGC variation between closely related strains from geographically distant environments. Results showed the greatest biosynthetic differences to be in a psychrotolerant Streptomyces strain, as well as a rare Actinobacteria genus, Kribbella, while two other Streptomyces spp. were surprisingly similar to known genomes. Streptomyces and Kribbella BGCs were predicted to encode antitumour, antifungal, antibacterial and biosurfactant‐like compounds, and the synthesis of NPs with antibacterial, antifungal and surfactant properties was confirmed through bioactivity assays.

Friday 4 September 2020

We are recruiting! Two year Postdoc available in plant conservation genomics

We have a two year full time postdoc position available to conduct bioinformatics, laboratory and field research in the area of conservation genomics as part of ARC Linkage Project LP180100721, and assist in the supervision of honours and postgraduate research students as required. This is a collaborative project between the University of New South Wales, Royal Botanic Gardens Sydney and the Australian National University to develop approaches for the conservation of plant species that are threatened by a fungal pathogen (Austropuccinia psidii, the cause of myrtle rust). Techniques involved include but are not limited to field collection of plant material, DNA extraction, plant growth experiments, execution of software for bioinformatic and statistical analyses.

More details can be found on the UNSW jobs site here: Informal enquiries are welcome if you want to know more about the project.

Wednesday 22 July 2020

Computational Prediction of Disordered Protein Motifs Using SLiMSuite

Edwards RJ, Paulsen K, Aguilar Gomez CM & Pérez-Bercoff Å (2020): Computational Prediction of Disordered Protein Motifs using SLiMSuite. Methods Mol Biol. 2141:37-72. doi: 10.1007/978-1-0716-0524-0_3. [PubMed]


Short linear motifs (SLiMs) are important mediators of interactions between intrinsically disordered regions of proteins and their interaction partners. Here, we detail instructions for the computational prediction of SLiMs in disordered protein regions, using the main tools of the SLiMSuite package: (1) SLiMProb identifies and calculates enrichment of predefined motifs in a set of proteins; (2) SLiMFinder predicts SLiMs de novo in a set of proteins, accounting for evolutionary relationships; (3) QSLiMFinder increases SLiMFinder sensitivity by focusing SLiM prediction on a specific query protein/region; (4) CompariMotif compares predicted SLiMs to known SLiMs or other SLiM predictions to identify common patterns. For each tool, command-line and online server examples are provided. Detailed notes provide additional advice on different applications of SLiMSuite, including batch running of multiple datasets and conservation masking using alignments of predicted orthologues.

Tuesday 23 June 2020

Do epigenetic changes drive corticosterone responses to alarm cues in larvae of an invasive amphibian?

Our latest cane toad paper is now online at Integrative and Comparative Biology, using our draft cane toad genome as the reference for differential methylation analysis. (Thanks to coronavirus delays, the updated cane toad genome was not quite ready for this analysis but watch this space!)

Sarma RR, Edwards RJ, Crino OL, Eyck HJF, Waters PD, Crossland MR, Shine R & Rollins LA (accepted): Do epigenetic changes drive corticosterone responses to alarm cues in larvae of an invasive amphibian? Integrative and Comparative Biology icaa082


The developmental environment can exert powerful effects on animal phenotype. Recently epigenetic modifications have emerged as one mechanism that can modulate developmentally plastic responses to environmental variability. For example, the DNA methylation profile at promoters of hormone receptor genes can affect their expression and patterns of hormone release. Across taxonomic groups, epigenetic alterations have been linked to changes in glucocorticoid (GC) physiology. GCs are metabolic hormones that influence growth, development, transitions between life-history stages, and thus fitness. To date, relatively few studies have examined epigenetic effects on phenotypic traits in wild animals, especially in amphibians. Here, we examined the effects of exposure to predation threat and experimentally manipulated DNA methylation on corticosterone (CORT) levels in tadpoles and metamorphs of the invasive cane toad (Rhinella marina). We included offspring of toads sampled from populations across the species’ Australian range. In these animals, exposure to chemical cues from injured conspecifics induces shifts in developmental trajectories, putatively as an adaptive response that lessens vulnerability to predation. We exposed tadpoles to these alarm cues, and measured changes in DNA methylation and CORT levels, both of which are mechanisms that have been implicated in the control of phenotypically plastic responses in tadpoles. To test the idea that DNA methylation drives shifts in GC physiology, we also experimentally manipulated methylation levels with the drug zebularine. We found differentially methylated regions between control tadpoles and their full-siblings exposed to alarm cues, zebularine or both treatments. However, the effects of these manipulations on methylation patterns were weaker than clutch (e.g. genetic, maternal, etc.) effects. CORT levels were higher in larval cane toads exposed to alarm cues and zebularine. We found little evidence of changes in DNA methylation across the glucocorticoid receptor gene (NR3C1) promoter region in response to alarm cue or zebularine exposure. In both alarm cue and zebularine-exposed individuals, we found differentially methylated DNA in the suppressor of cytokine signaling 3 gene (SOCS3), which may be involved in predator avoidance behavior. In total, our data reveal that alarm cues have significant impacts on tadpole physiology, but show only weak links between DNA methylation and CORT levels. We also identify genes containing differentially methylated regions in tadpoles exposed to alarm cues and zebularine, particularly in range-edge populations, that warrant further investigation.

Monday 1 June 2020

Johansson SA, Stephenson P, Edwards RJ, Yoshida K, Moore M, Terauchi R, Zubkov MV, Terry MJ & Bibby TS (2020): Isolation and molecular characterisation of Dunaliella tertiolecta with truncated light-harvesting antenna for enhanced photosynthetic efficiency. Algal Research 48:101917.


  • Liquid based screening and selection of algae with truncated light harvesting antenna
  • Selection of algae with improved photosynthetic efficiency
  • Genome independent transcriptome profiling characterises photosynthetic physiology.
  • Develops potential to exploit the natural diversity of microalgae for biotechnology


Here we report the development of a high-throughput selection protocol using random mutagenesis and live single-cell sorting to isolate cell lines from the algae Dunaliella tertiolecta with reduced chlorophyll content, with the aim to optimise the antenna size for increased photosynthetic efficiency. Two promising cell lines (lca1 and lca2) have been isolated that display a truncated light-harvesting antenna, and hence improved photosynthetic energy conversion efficiency by increasing the light intensity at which photosynthesis becomes saturated (Is). lca1 and lca2 differ significantly: the lca2 phenotype retains an ability to alter its antenna size in response to varying light intensity, whereas lca1 appears to have lost this ability and is ‘locked’ to a truncated antenna and high-light phenotype. Despite these clear differences, transcriptomic analysis shows that the expression profiles for differentially expressed nuclear-encoded photosynthetic genes is similar in both lca1 and lca2, possibly suggesting underlying mutations in the regulation of photosynthesis are causing the observed changes in phenotype rather than mutations impacting specific components of the photosynthetic apparatus. The combination of approaches presented here offer the capacity to substantially improve photosynthetic efficiency from any microalgal species irrespective of the extent to which it has been characterised genetically or the availability of molecular tools for rational engineering. It thus offers the potential to begin to exploit the huge natural diversity of microalgae for enhanced biomass production.

Thursday 2 April 2020

Canfam_GSD: De novo chromosome-length genome assembly of the German Shepherd Dog (Canis lupus familiaris) using a combination of long reads, optical mapping, and Hi-C

Our latest paper is out! This one is a bit more photogenic than the cane toad - a German Shepherd Dog called Nala. This was a big international effort in a collaboration led by Bill Ballard at UNSW that included a dozen institutions across four continents. We threw all the main sequencing technologies at this one and achieved a chromosome-level assembly of better quality than the current “CanFam” reference genome.

You can find out more in the UNSW press release.

Field MA, Rosen BD, Dudchenko O, Chan EKF, Minoche AM, Edwards RJ, Barton K, Lyons RJ, Enosi Tuipulotu D, Hayes VM, Omer AD, Colaric Z, Keilwagen J, Skvortsova K, Bogdanovic O, Smith MA, Lieberman Aiden E, Smith TPL, Zammit RA & Ballard JWO (2020): Canfam_GSD: De novo chromosome-length genome assembly of the German Shepherd Dog (Canis lupus familiaris) using a combination of long reads, optical mapping, and Hi-C. GigaScience 9(4):giaa027. [GigaScience]



The German Shepherd Dog (GSD) is one of the most common breeds on earth and has been bred for its utility and intelligence. It is often first choice for police and military work, as well as protection, disability assistance, and search-and-rescue. Yet, GSDs are well known to be susceptible to a range of genetic diseases that can interfere with their training. Such diseases are of particular concern when they occur later in life, and fully trained animals are not able to continue their duties.


Here, we provide the draft genome sequence of a healthy German Shepherd female as a reference for future disease and evolutionary studies. We generated this improved canid reference genome (CanFam_GSD) utilizing a combination of Pacific Bioscience, Oxford Nanopore, 10X Genomics, Bionano, and Hi-C technologies. The GSD assembly is ∼80 times as contiguous as the current canid reference genome (20.9 vs 0.267 Mb contig N50), containing far fewer gaps (306 vs 23,876) and fewer scaffolds (429 vs 3,310) than the current canid reference genome CanFamv3.1. Two chromosomes (4 and 35) are assembled into single scaffolds with no gaps. BUSCO analyses of the genome assembly results show that 93.0% of the conserved single-copy genes are complete in the GSD assembly compared with 92.2% for CanFam v3.1. Homology-based gene annotation increases this value to ∼99%. Detailed examination of the evolutionarily important pancreatic amylase region reveals that there are most likely 7 copies of the gene, indicative of a duplication of 4 ancestral copies and the disruption of 1 copy.


GSD genome assembly and annotation were produced with major improvement in completeness, continuity, and quality over the existing canid reference. This resource will enable further research related to canine diseases, the evolutionary relationships of canids, and other aspects of canid biology.

Photo credit: Outdoor Action Photography.

Tuesday 18 February 2020

Jack Clarke (Honours student)

Jack worked in the Edwards Lab in 2020 for his Honour years, researching the evolution of snake venoms using the lab’s de novo genome assemblies of two Australian snakes (Eastern brown snake and mainland tiger snake). His researched focused on phylogenetic analysis of venom gene families in order to understand how venom divergence has occurred in different branches of the snake lineage. The work focused on the role that tandem gene duplication events had played in the evolution of specific venom gene families and identified novel sites of tandem duplication shared across multiple snake species.

Jack holds a Bachelor of Advanced Science (Honours) (First Class) with majors in Bioinformatics and Genetics. He is currently completing his PhD at UNSW in collaboration with the Victor Chang institute.


Friday 31 January 2020

Research snapshot - January 2020

One of the most important, interesting and challenging questions in biology is how new traits evolve at the molecular level. My lab employs sequence analysis techniques to interrogate DNA and protein sequences for the signals left behind by evolution. We are a bioinformatics lab but like to incorporate bench/field data through collaboration wherever possible.

Main Research

Building on a solid foundation of bioinformatics and evolutionary theory, we apply genomics, transcriptomics, proteomics and interactomics and systems analysis to understand complex biological systems. Core research activities can be broadly divided into two main themes:

  1. Evolutionary Genomics, with a focus on applying de novo genome assembly and population genomics to problems in ecology and biotechnology.

  2. Protein-protein interactions, with a focus on the prediction of short linear interaction motifs and their role in human health and disease, including host-pathogen interactions.

These are explored in more detail, below.

1. Evolutionary Genomics.

The main research focus of the lab is the exploitation of genomic and post-genomic data to understand biological function and adaptation to novel environments. We work closely with the Ramaciotti Centre for Genomics and are involved in numerous de novo whole genome sequencing and assembly projects, using short read (Illumina), long read (PacBio & Nanopore) and linked read (10x Chromium) sequencing. One of the biggest of these is leading the bioinformatics and assembly effort in a consortium to sequence the cane toad genome, and leading the BABS Genome project to sequence two iconic Australian snakes. We are a member of the Oz Mammals Genomics initiative, assisting with the sequencing and assembly of Australia’s unique marsupial fauna. In 2018, we were selected as part of a team to sequence the Waratah genome as part of the pilot phase for the new Genomics of Australian Plants initiative.

We enjoy bringing our bioinformatics to bear on a variety of collaborative research projects. Most notably, we have an ARC Linkage Grant with Microbiogen Pty Ltd to understand how a strain of Saccharomyces cerevisiae has evolved to efficiently use xylose as a sole carbon source: something vital for second generation biofuel production that wild yeast cannot do. We are combining comparative genomics, evolutionary genetics, RNA-Seq transcriptomics, and competition assays to understand how the novel metabolism evolved. Through deep Illumina resequencing of evolving populations, and assembling reliable complete genomes of the founding ancestors, the ultimate goal is to trace how mutations have interacted with existing genetic variation during adaptive evolution. More recently, we have received an ARC Linkage Grant with the Royal Botanic Gardens and Domain Trust, to apply genomics approaches to the challenges of rainforest tree conservation in the face of climate change and invasive pathogens. We are also collaborating with industrial and academic partners to de novo sequence, assemble, annotate and interrogate the genomes of a selection of microbes with interesting metabolic abilities.

2. Short Linear Motifs (SLiMs).

Many protein-protein interactions are mediated by Short Linear Motifs (SLiMs): short stretches of proteins (5-15 amino acids long), of which only a few positions are critical to function. These motifs are vital for biological processes of fundamental importance, acting as ligands for molecular signalling, post-translational modifications and subcellular targeting. SLiMs have extremely compact protein interaction interfaces, generally encoded by less than 4 major affinity-/specificity-determining residues. Their small size enables high functional density and evolutionary plasticity, making them frequent products of convergent “ex nihilo” evolution. It also makes them challenging to identify, both experimentally and computationally.

A major focus of the lab is the computational prediction of SLiMs from protein sequences. This research originated with Rich’s postdoctoral research, during which he developed a sequence analysis methods for the rational design of biologically active short peptides. He subsequently developed SLiMDisc, one of the first algorithms for successfully predicting novel SLiMs from sequence data - and coined the term “SLiM” into the bargain. This subsequently lead to the development of SLiMFinder, the first SLiM prediction algorithm able to estimate the statistical significance of motif predictions. SLiMFinder greatly increased the reliability of predictions. SLiMFinder has since spawned a number of motif discovery tools and webservers and is still arguably the most successful SLiM prediction tool on benchmarking data. Methods are made available through the SLiMSuite bioinformatics package and webservers.

Current research is looking to develop these SLiM prediction tools further and apply them to important biological questions. Of particular interest is the molecular mimicry employed by viruses to interact with host proteins and the role of SLiMs in other diseases, such as cancer. Other work is concerned with the evolutionary dynamics of SLiMs within protein interaction networks.


In addition to the main research in the lab, the lab has a number of interdisciplinary collaborative projects applying bioinformatics tools and molecular evolution theory to experimental biology, often using large genomic, transcriptomic and/or proteomic datasets. These projects often involve the development of bespoke bioinformatics pipelines and a number of open source bioinformatics tools have been generated as a result. Please see the Publications and Lab software pages for more detail, or get in touch if something catches your eye and you want to find out more. We frequently have small collaborations and/or undergraduate student research projects. Many of these are “on hold” waiting for the right person, or sometimes data, to come along. If you think that you have what it needs, get in touch!