InCoB2014: mining biological data from genomics for transforming industry and health

Abstract

The 13th International Conference on Bioinformatics (InCoB2014) was held for the first time in Australia, at Sydney, July 31-2 August, 2014. InCoB is the annual scientific gathering of the Asia-Pacific Bioinformatics Network (APBioNet), hosted since 2002 in the Asia-Pacific region. Of 106 full papers submitted to the BMC track of InCoB2014, 50 (47.2%) were accepted in BMC Bioinformatics, BMC Genomics and BMC Systems Biology supplements, with three papers in a new BMC Medical Genomics supplement. While the majority of presenters and authors were from Asia and Australia, the increasing number of US and European conference attendees augurs well for the international flavour of InCoB. Next year's InCoB will be held jointly with the Genome Informatics Workshop (GIW), September 9-11, 2015 in Tokyo, Japan, with a view to integrate bioinformatics communities in the region.

Introduction

The 13th InCoB (International Conference on Bioinformatics), an official conference of the Asia-Pacific Bioinformatics Network (APBioNet) [1], was held in Sydney, Australia. In keeping with InCoB's tradition to locate the conference in different Asia-Pacific countries, InCoB went to Australia, for the first time, with the NSW Chief Scientist and Engineer, Professor Mary O'Kane giving the opening address. The conference was well attended by the local community as well as participants from Asia-Pacific, Europe and USA, with scientists from Qatar and Turkey attending InCoB for the first time. The plenary talks covered mutations in DNA, proteomics, RNA Biology, systems biology, mathematics, statistics and computer science as well as bioinformatics training to equip scientists with the current "tools of the trade" and best practice methodologies, while the conference attendees presented cutting edge research with three industry presentations, 89 orals as well as 45 posters.

Manuscript submission and review

We offered authors four tracks to submit manuscripts for potential publication in the supplement issues of BMC Bioinformatics, BMC Systems Biology or BMC Genomics (BMC track) and PeerJ [2]. Of the 112 submitted manuscripts, 106 were in the BMC track. All manuscripts received at least two reviews from the 92 member Program Committee, supported by 35 additional reviewers (Additional file 1). The first round of reviews resulted in the provisional acceptance of 12 (11.3%) manuscripts, with minor revisions. The authors of 45 manuscripts, including three that were transferred by the Program Committee Co-chairs to the PeerJ track, had to address major concerns raised by the reviewers. After a second round of review, another 38 (35.8%) manuscripts were provisionally accepted pending minor revisions. In all cases, the reviewers assessed the manuscripts at least twice and all submissions were ranked based on the reviewers' scores, in accord with earlier InCoB publications [3].

The 20 articles in this supplement cover mainly "genomic" topics, with three medically-oriented papers going into BMC Medical Genomics [4] for the first time. The InCoB2014 BMC Bioinformatics supplement [5] comprises 16 manuscripts while another 11 papers are presented in the BMC Systems Biology supplement [6]. Five more submissions have been published in the InCoB2014 PeerJ collection [7].

Sequencing, genomes and genome analysis

With genome sequencing technologies becoming more and more accessible and affordable, the genetic origin of the Marwari horse was established by whole genome sequencing [8] while the chloroplast genome of Australia's macadamia nut tree [9]. funRNA [10] is a collection of RNA interference (RNAi) genes, implicated in genome defence as well as diverse cellular, developmental, and physiological processes, from fungal, metazoan and plant genomes along with bacterial and archaeal genomes. Abbas et al.[11] have assessed de novo assembly software for fungal genome data. Proteogenomics is increasingly used for the accurate annotation of protein coding regions using proteomic data. As currently available proteogenomic tools are tailored specifically for human and eukaryotic data, Uszkoreit et al.[12] have developed a proteogenomic analysis pipeline, specifically for bacterial genomes.

Transcriptomics

Seven papers are devoted to transcriptome analysis addressing a range of challenges from epigenetics [13, 14] to understanding transcript-level changes with diseases [15–18]. dCAP [13] is a new method to simultaneously detect constitutive and differential regulation of multiple epigenetic factors from multiple sample datasets, while the YNA database [14] provides an integrated data mining platform for chromatin changes in yeast. Huang et al.[15] have identified transcript-level changes implicated in biological dysfunction of energy metabolism and hemostasis in schizophrenia while Yarmishyn et al.[16] have pinpointed a non-coding RNA as a novel marker of neuroblastoma progression. Sheng et al.[17] have identified the most common microRNA (miRNA) editing events in colon cancer. For non-small cell lung cancer, Mah et al.[18] have identified a single single nucleotide polymorphism (SNP) as a good prognostic marker of patient outcome following chemotherapy. While transcripts are in the main used as a measure of protein expression, Sun et al.[19] have applied transcriptomics and pathway analysis to improve protein identification from proteomics data, especially where the sample size is limited to a few cells.

Functional genomics

Guanine-rich nucleotide sequences form four-stranded G-quadruplex structures. Yano and Kato [20] have used hidden Markov Models to reliably identify these structures, especially those involved in DNA transcription. Wu et al.[21] have developed a novel algorithm (GM-SMCC), for predicting the functional annotation of protein-coding genes, applied successfully to the 2001 KDD cup yeast gene datasets. Small ubiquitin-like modifier (or SUMO) proteins covalently attached other proteins lead to sumoylation, which is involved in various cellular processes, including nuclear-cytosolic transport, transcriptional regulation, apoptosis, protein stability, response to stress, and cell cycle progression. Yavuz and Sezerman [22] have developed an accurate support vector machine (SVM)-based approach to identify these sites from sequence data, as a precursor to experimental validation.

Pharmacogenomics

In the era of genomic medicine, it is possible that some approved drug molecules can be used for diseases other than those they were originally approved for. Yang and co-workers [23] propose the concept of "Homopharma", to combine similar drug binding environments to better understand molecular binding mechanisms for deploying approved drugs for other diseases, known as "repurposing." Tyagi et al.[24] have optimized arylthioindole compounds for efficiently disrupting tubulin assembly towards anti-cancer therapy, using QSAR and molecular dynamics approaches.

Disease informatics

Taguchi and co-workers [25] have identified TINAGL1 and B3GALNT1 as novel key candidate genes in the treatment of non-small cell lung cancer from gene expression and epigenetic data, while sets of microRNA biomarkers for detecting lung squamous cell carcinoma have been proposed by Song et al.[26]. Xu et al.[27] have developed a novel methodology, MHC2MIL for developing peptide-based vaccines, benchmarked on 12 HLA DP and DQ molecules.

Medical informatics

Papers specifically relating on genome-scale analysis with a disease focus are presented in a new supplement in BMC Medical Genomics and a brief overview of these articles is presented here. With pandemic viral infections spreading rapidly by jet travel, understanding cross-species transmissibility of vectors is addressed by Tan and co-workers [28] for influenza A, using a random forest approach. In the quest for diagnostic biomarkers and therapeutic targets, Nenadic and co-workers [29] have developed a novel text-mining approach to successfully identify novel genes and pathways for thyroid cancer subtypes. While gene-based cancer biomarkers are typically sets of hundreds or thousands of genes, Olsen et al.[30] have analysed public proteogenomic data to zoom in on just 32 tumor antigenic proteins as biomarkers for invasive ductal carcinomas. These studies point to the combined use of several "-omic" technologies as the focus of future studies for understanding, detecting and combatting diseases.

Conclusion

With the growth in regional bioinformatics meetings, including ISCB-Asia meetings and the 2013 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) in China [31], there is a growing community of bioinformatics scientists in the Asia-Pacific. To consolidate multiple meetings and to provide cross-talk between traditionally different bioinformatics communities, we invite you to attend the 2015 InCoB meeting to be held jointly with the Genome Informatics Workshop (GIW) in Tokyo, Japan [32].