Archives: Projects Blog

 … Peter Georgeson, Anna Syme, Clare Sloggett, Jessica Chung, Harriet Dashnow, Michael Milton, Andrew Lonsdale, David Powell, Torsten Seemann, Bernard Pope

GigaScience, Volume 8, Issue 9

Bioinformatics software tools are often created ad hoc, frequently by people without extensive training in software development. In particular, for beginners, the barrier to entry in bioinformatics software development is high, especially if they want to adopt good programming practices. Even experienced developers do not always follow best practices. This results in the proliferation of poorer-quality bioinformatics software, leading to limited scalability and inefficient use of resources; lack of reproducibility, usability, adaptability, and interoperability; and erroneous or inaccurate results.

Bionitio is a tool that automates the process of starting new bioinformatics software projects following recommended best practices. With a single command, the user can create a new well-structured project in 1 of 12 programming languages. The resulting software is functional, carrying out a prototypical bioinformatics task, and thus serves as both a working example and a template for building new tools. Key features include command-line argument parsing, error handling, progress logging, defined exit status values, a test suite, a version number, standardised building and packaging, user documentation, code documentation, a standard open source software license, software revision control, and containerisation.

Recommended:

• for beginner-to-intermediate bioinformatics programmers
• as an excellent starting point for new projects
• to help developers adopt good programming practices from the beginning of a project
• for encouraging high-quality tools to be developed more rapidly
• for enabling tools to be more easily installed and used in a consistent way.

Bionitio is released as open source software under the MIT License and is available at https://github.com/bionitio-team/bionitio.

Congratulations to a great team from Melbourne Bioinformatics and close colleagues in our networks who collaborated on this truly community-focussed project.

This story was first published in September 2019.

2020 highlights

The project has progressed well in its development of a method to predict time of cycle in endometrium expression data and normalisation. How this is applied clinically is yet to be clearly articulated but it is proving of interest to the field.

Publications

Holdsworth-Carson SJ, Chung J, Sloggett C, Mortlock S, Fung JN, Montgomery GW, Dior UP, Healey M, Rogers PA, & Girling JE (2020). Obesity does not alter endometrial gene expression in women with endometriosis. Reproductive Biomedicine Online.

Another paper has been accepted for publication in 2021.

Grants

In 2020, the project team was awarded a Medical Research Future Fund (MRFF) grant for the Royal Women’s Hospital for improving diagnosis and treatment of endometriosis.

2020 highlights

• Chol-Hee Jung has been managing Australian data locally and in charge of organising them within the European Genome-phenome Archive (EGA) for sharing across the Pan Prostate Cancer Group (PPCG).
• He has also processed and been processing Australian data through various pipelines for the identification of genomic, epigenomic and transcriptomic variants.
• As part of the study within PRECEPT and/or PPCG, analyses are focussed not only on individual ‘omics data but also associations between them.
• From the collaboration within PPCG, two manuscripts have been submitted for publication, and two are under preparation.
• Genomic variants identification pipelines developed by international collaborators within PPCG are managed by Chol-Hee who has also been developing an R package (interim name ‘dmplorer’) which helps visualise DNA methylation microarray data analysis results.

Grants

MEMENTO: Biomarker discovery in metastatic hormone sensitive prostate cancer”. Below The Belt Research Fund, Australia and New Zealand Urogenital and Prostate Cancer Trials Group. $50,000 AUD direct. Anis Hamid (PI), Ian Davis, Christopher Sweeney, Arun Azad, Ben Tran, Shahneen Sandhu, Chris Hovens, Niall Corcoran, Bernard Pope, Chol-hee Jung, Yi Sun.

Khalid Mahmood continued to co-supervise 3^rd year PhD student Philip Harraka with Prof Judy Savige (RMH).

Publications in preparation

Community Engagement

Presentations

The Portable Pipelines Project started the development of Janis in late 2018 with the goal of delivering bioinformatics workflows that are portable and reproducible across many compute environments. Janis is a Python framework that provides a simple Application Programming Interface to build and run workflows that adhere to current standards of workflow specifications.

Over 2020, the team focused on expanding Janis’ community by running workshops, developing new pipelines and introducing more features to ease new users’ onboarding into the ecosystem.

2020 developments

• The team developed workshop materials delivered at the Peter Mac workshops and the Bioinformatics Community Conference (BCC) 2020. In total, more than 50 participants have joined Janis workshops across these two events.
• Janis has been integrated into Peter Mac automated bioinformatics analysis system, responsible for analysing hundreds of cancer sequencing samples every week across research and molecular pathology.
• The number of tools and examples available in janis-bioinformatics has grown to more than 100 modules, with contributions from various institutes in Australia and overseas.
• Introduced a new Janis development toolkit that enables importing of existing CWL pipelines into Janis, automated reference genome downloads, input validation and continuous integration tests of bioinformatics tools.

Community uptake

The project has continued its contribution to open source repositories including CWL, OpenWDL and Cromwell. In addition, Janis has now been included in the list of converters and code generators for CWL .

The project has also seen an increased uptake in cancer bioinformatics applications. A project involving multiple tumours from the same individual (CASCADE project) was recently analysed with workflows written in Janis (and this work has been submitted for publication). Melbourne Bioinformatics senior bioinformatician Assoc Prof Bernard Pope reports using Janis in his own work on Prostate Cancer through the PRECEPT project and a project investigating brain cancer with collaborators from Royal Melbourne Hospital, the Peter Doherty Institute and WEHI.

In 2020, Janis also welcomed a new collaboration with the Australian BioCommons, with a commitment to extend Janis’ use across the Australian research community.

Presentations

• Bioinformatics Community Conference (BCC) 2020
• WEHI seminar series
• Monash Bioinformatics Platform Seminar
• Somatic variant calling in whole genome sequencing, opportunities, challenges and solutions, Monash Partners Comprehensive Cancer Consortium, Tech Talks series, 20 November 2020
• The Australian BioCommons Webinar series

Upcoming events

In 2021, more workshops on Janis will be scheduled to be run as part of the Parkville Bioinformatics Training Group’s activities led out of Melbourne Bioinformatics. This work will also extend beyond through the Australian BioCommons training program.

What’s next for Janis?

In 2021, the team looks to support more researchers to use Janis to analyse their data and build new workflows. The team aims to add new support for other workflow specifications, such as Nextflow, and continue to build more exemplar pipelines targeting other domains beyond cancer bioinformatics.

Large datasets generated by researchers using liquid chromatography-mass spectrometry to identify unknown drug metabolites can now be processed more quickly and efficiently by using HiTIME, a novel memory efficient and scalable parallelisation algorithm which allows timely processing  on commodity computing hardware.

Computer scientists and chemists from the University of Melbourne and University of NSW have now published a paper documenting their successful work in developing this algorithm at Software X, Elsevier’s open access journal: Michael G. Leeming, Andrew P. Isaac, Luke Zappia, Richard A.J. O’Hair, William A. Donald and Bernard J. Pope, HiTIME: An efficient model-selection approach for the detection of unknown drug metabolites in LC-MS data.

The identification of metabolites plays an important role in understanding drug efficacy and safety however these compounds are often difficult to identify in complex mixtures. One approach to identify drug metabolites involves utilising differentially isotopically labelled drug compounds to create unique isotopic signals that can be detected by liquid chromatography-mass spectrometry (LC-MS). User-friendly, efficient, computational tools that allow selective detection of these signals are lacking. Our computer scientists have developed an efficient open-source software tool called HiTIME (High-Resolution Twin-Ion Metabolite Extraction) which filters twin-ion signals in LC-MS data.

HiTIME is a sensitive tool for the detection of twin-ion signals in LC-MS data that has been successfully demonstrated for the detection of paracetamol (APAP) metabolites in blood plasma of APAP-treated rats and endogenous proteins covalently bound to electrophilic APAP metabolites. HiTIME accepts inputs and produces outputs in standard mzML format, facilitating integration with other tools and workflows. A significant advantage of HiTIME is that it supports inputs in both profile and centroid modes, and its novel memory efficient and scalable parallelisation algorithm allows timely processing of large data sets on commodity computing hardware.

Image source: Wikipedia

This work was completed with support from a University of Melbourne Interdisciplinary Seed Grant, the Victorian Life Sciences Computation Initiative (now Melbourne Bioinformatics), Assoc. Prof. Pope’s Victorian Health and Medical Research Fellowship, Australia and Dr Michael Leeming’s Elizabeth and Vernon Puzey PhD scholarship and The University of Melbourne’s Norma Hilda Schuster scholarship.

Two weeks ago, members of Melbourne Bioinformatics (Andrew Lonie (Director Australian BioCommons) and Simon Gladman) contributed to an extraordinary open science effort to pull together all the current public genomic data on SARS-CoV-2, in demonstration of a collaborative, globally accessible, rapid, reproducible research response to this current public health crisis. It’s a wonderful example of the benefit of shared bioinformatics infrastructure for medical research.

No more business as usual: agile and effective responses to emerging pathogen threats require open data and open analytics (link to BioRxiv paper)

Abstract
The current state of much of the Wuhan pneumonia virus (COVID-19) research shows a regrettable lack of data sharing and considerable analytical obfuscation. This impedes global research cooperation, which is essential for tackling public health emergencies, and requires unimpeded access to data, analysis tools, and computational infrastructure. Here we show that community efforts in developing open analytical software tools over the past ten years, combined with national investments into scientific computational infrastructure, can overcome these deficiencies and provide an accessible platform for tackling global health emergencies in an open and transparent manner. Specifically, we use all COVID-19 genomic data available in the public domain so far to (1) underscore the importance of access to raw data and to (2) demonstrate that existing community efforts in curation and deployment of biomedical software can reliably support rapid, reproducible research during global health crises.

Their draft paper concludes:
“… anyone can use the open workflows described here to analyze the new data. In an age of digital connectedness, open, highly accessible, globally shared data and analysis platforms have the potential to transform the way biomedical research is done, opening the way to ‘global research markets’, where competition arises from deriving understanding rather than access to samples and data. Other disciplines have embraced the benefits of global data generation and sharing, astronomy and high energy physics being two highly successful examples. We have the opportunity to mirror their successes in infrastructure funding by demonstrating that biological research can embrace the same global perspective on common infrastructure investment and data sharing…”

This resource is also empowering our global Galaxy community (we’ve run workshops in Africa, SE Asia, EU, USA, AU) and our students who have trained with us on the Galaxy platform who can now get in and do their own research.

Galaxy Australia is hosted by the University of Melbourne and the Queensland Cyber Infrastructure Foundation and funded through NCRIS. It is one of the 4 global Galaxy platforms that participated in the project. 

Researchers across the Parkville Precinct and beyond now have a simpler way to run their data analysis pipelines on multiple computing platforms thanks to recent developments in the Portable Pipelines Project.

Over 2019, our team of experts across WEHI, PeterMac and Melbourne Bioinformatics built Janis, a new Python framework for building and running workflows. Janis provides a simple Application Programming Interface to build workflows that are portable and reproducible across many compute environments and it provides workflow specifications (Common Workflow Language [CWL] and Workflow Description Language [WDL]) as publishable artifacts.

2019 developments

• Janis’ portability has been validated on systems at Peter Mac, WEHI, University of Melbourne, Pawsey Supercomputing Centre in WA and the Google Cloud Platform.
• The project has generated considerable community awareness and support on GitHub and other online platforms.
• Janis is in alpha testing now and is openly available for researchers to download and use through GitHub (support is available through GitHub and Gitter).
• It is being successfully used by researchers at Peter Mac, WEHI and Melbourne Bioinformatics to analyse a mixture of whole genome and exome sequencing data.
• Documentation now available here.

Community uptake

Uptake across the life science community has further validated Janis’ usefulness, with Kersten Breuer from the German Cancer Research Center adding support for Janis workflows to CWLab. And at a combined Broad Institute and AWS hackathon Janis was adopted to improve cloud support for Cromwell.

The project has now contributed to open source repositories including the CWL, WDL, Cromwell and more, and in particular the team has worked to document Singularity support for Cromwell.

Presentations

• Bioinformatics Open Source Conference (BOSC) 2019 – Basel
• Victorian Bioinformatics Seminar (VCBS) 2019 – Melbourne
• Australian Bioinformatics and Computational Biology Society (ABACBS) 2019 – Sydney
• WEHI seminar series – Melbourne
• Australian BioCommons briefing online across Australia

Upcoming events

In early 2020, workshops on Janis will run as part of the Parkville Bioinformatics Training Group’s activities led out of Melbourne Bioinformatics.

What’s next for Janis?

In 2020 the team is looking to support more researchers to use Janis to analyse their data and build new workflows. The team aims to build exemplar pipelines for analysing RNA seq data, run workshops, complete documentation and work closely with researchers to increase functionality.

This has been an exciting and productive collaboration to date. And given Janis is applicable for all research domains, it will also be interesting to see how it develops a life beyond its original intended use.

2019 update:

Chol Hee Jung has been carrying out the quality control and processing of Australian data and the identification and analysis of genomic variants using various analysis pipelines on this major research project. He also handled the local management of data and organised data sharing with international collaborators.

Further funding

Preliminary investigations have contributed to the award of $4million over 3 years for the Australian PRECEPT Program funded by the Prostate Cancer Research Alliance, an Australian Government and Movember Foundation Collaboration. Lead investigator is A/Prof Niall Corcoran and Chief investigators include A/Prof Bernie Pope and A/Prof Danny Park from Melbourne Bioinformatics.

Project description

This project aims to reveal how the tumour progresses to lethal metastatic stages and the detailed view of tumour cells by integrated genomic and epigenomic variants analyses from cohort patients. A part of the work from this project also contributes to the international Pan-Prostate Cancer Group collaboration.

Project collaborators

Grants

2019 update:

This work progressed well in 2019, with Gayle Philip and Dieter Bulach sharing their expertise with the Prof van Oppen’s team.

Along with conducting her own analysis of high-throughput data generated by the lab, Gayle has been upskilling members of the lab to be able to perform their own analyses. This has included implementing systems for storage of data in MediaFlux, communication in the lab through MS Teams and teaching lab members how to access the University of Melbourne’s High Performance Computing system (Spartan). Through the lab’s association with the Environmental Microbiology Research Initiative (EMRI), Gayle has delivered workshops for EMRI including: Galaxy and Data Formatting, Nectar and Spartan HPC and Introduction to Unix.

Project description

This research focuses on microbial symbiosis in corals, adaptation/acclimatisation to climate change, and connectivity of coral reefs. It is particularly focussed on ‘assisted evolution’, where mechanisms of adaptation and acclimatisation in corals and genetic manipulations to enhance stress tolerance and fitness of corals in a changing environment are explored.

Read more about Prof van Oppen’s work here.

Project collaborators

Grants

2019 update:

This project continued to progress well in 2019, with Khalid Mahmood sharing his expertise with the group.

Several significant collaborations have developed across several projects in the laboratory. The focus of these collaborations has been to use genomics and associated clinical data to characterise CRCs to improve screening and diagnostics strategies for patients. Some of the key tasks have been to deploy state of the art bioinformatics methods to analyse germline and tumour genomics sequencing data to characterise different subtypes of CRCs as well as identity new variations and genes that predispose families to higher risk of developing CRC. Work from these collaborations has resulted in several publications under preparation or review. In addition, work has involved supervision of several honours students.

Publications

Tumor mutational signatures in sebaceous skin lesions from individuals with Lynch syndrome, Georgeson et al, Molecular Genetics and Genomic Medicine.

sEst: Accurate Sex-Estimation and Abnormality Detection in Methylation Microarray Data, Jung et al, International Journal of Molecular Science.

Presentations

Seminar, University of Melbourne Centre for Cancer Research

Oral presentation, International Conference InSiGHT (International Society for Gastrointestinal Hereditary Tumours).

Project description

The focus of the Colorectal Oncogenomics Group (COG) led by Assoc Prof Daniel Buchanan includes the identification and investigation of clinically and biologically relevant subtype of colorectal cancers (CRC) in both familial and non-familial settings. The analysis involves a wide range of multi-disciplinary techniques ranging from computational biology, epigenetics and genomics to analyse tumour and pre-malignant lesions in terms of their histopathological features. This very successful collaboration covers a range of colorectal cancer projects including those forming part of the University-hosted NHMRC Centre for Research Excellence in Optimising Screening for Colorectal Cancer, whose vision is to create and implement a personalised approach to colorectal cancer screening to reduce the number of new cases and deaths from this common disease.

Project collaborators

Grants

2019 update:

This project has made some very good progress in 2019, with Jessica Chung providing the expertise from our team to enable

•  development of a method to normalise cycle stage effects in endometrium expression data
•  developed an interactive R Shiny application where the research group can explore microarray and RNA-seq data with their own parameters
•  analysis of endometriosis severity and BMI, lipidomics data, uterine receptivity, and clinical factors that influence repeat surgery.

Project description

Endometriosis is a disorder that affects 5 – 10% of reproductive age women in Australia, causing severe pain and infertility. This project aims to use genomic data to identify candidate genes that increase the risk of endometriosis. We are also investigating mechanisms that cause reduced endometrial receptivity, the association between BMI and endometriosis, and clinical indicators that can predict repeat surgery for endometriosis.

Project collaborators

Grant

It’s one thing to identify a genetic disorder, but another to successfully treat it. In great news this month, Melbourne Genomics shared the story of two brothers whose genetic disorder was identified and treated, here in Melbourne. Thanks to all the teams involved, including our own team working on the genomic data analysis pipelines.

Full story.

Hi-Plex was developed by our Molecular Biologist, Assoc Prof Daniel Park and Computer Scientist, Assoc Prof Bernard Pope, co-leads of our Human Genomics Group at Melbourne Bioinformatics, to simplify processes and reduce costs on projects needing targeted sequencing of panels of genes across large numbers of specimens. It brings greater efficiency and accuracy to all such research projects – big and small.

Go here for background to the original Hi-Plex. 

25 February 2019

Following our July 2018 update, Genovic, Victoria’s shared clinical system for genomics, has now reached a further milestone. Last week Melbourne Genomics announced that GenoVic is now ready for use in clinical genomic testing.

Victorian Clinical Genetics Services (VCGS) is the first laboratory to agree to use the GenoVic system as its primary tool for genomic interpretation. The system will support VCGS’ medical scientists to identify and report on the exact changes in a patient’s DNA driving their condition or targetable for treatment.

Further agreements on implementing GenoVic are in process with Alliance members Monash Health, AGRF and The Royal Melbourne Hospital.

Get the full story at the Melbourne Genomics website.

Congratulations to Anthony Marty and his team on their contribution to the success of this project.