Our group expertise is in computational protein sequence and structure analysis to predict various aspects of
molecular and cellular functions (enzymatic activities, posttranslational modifications, cleavage, translocation
signals, 3D structures, effects of mutations, phylogenetic relationships, cellular pathways etc.) for discovering the
molecular mechanisms of biological and clinical phenotypes and experimental validation together with collaborators. Our
repertoire of computational analysis methods is applicable and useful in multiple research areas but our main focus
currently is on infectious diseases, human mutations, allergy and enzyme function prediction.
One of our traditional strongholds since the swine flu in 2009 is infectious disease research. Our FluSurver (http://flusurver.bii.a-star.edu.sg/) is the most complete one-stop influenza mutation analysis tool being used by researchers and surveillance experts globally. We have several published and ongoing projects with the WHO CC in Australia and National Influenza Centres relating to influenza drug resistance, viral fitness, host specificity and antigenic changes. The FluSurver is also a primary analysis tool for GISAID, the most complete influenza database also known for always hosting the latest outbreak sequences.
One of our highlights in influenza research in 2018 was a landmark publication in the journal Nature Ecology and
Evolution by our PhD student Alvin in collaboration with Colin Russell in Amsterdam. In our fully computational
study, we analyzed the genetic sequences of >25,000 influenza viruses collected between 2009 and 2016 as part
of global routine surveillance efforts. We investigated the mutational patterns between closely-related viruses using
the age of the infected hosts as a proxy of their influenza immune experience - older adults are more likely to have
prior infections of influenza compared to young children. We found that individuals with different infection histories
of influenza were frequently infected by genetically identical viruses. Virus mutations that arose in previously infected
adults were highly similar to those found in very young children who might not have been infected with flu before.
Importantly, mutations likely to cause antigenic change were rarely found and were not associated with the extent of the
expected influenza infection histories of individuals. In other words, our study suggests that individual immune selection
only plays a limited role in influenza virus evolution.
Because we can quickly go from genomes to protein structures through modelling in our computers often only
requiring the new sequences as input, our group offers powerful support in infectious disease surveillance and rapid outbreak investigations to get a quick handle on bugs here and around the world. Besides Influenza, we also helped
characterizing MERS, Ebola, HIV, Noro, Adeno, Hepatitis C, West Nile, Dengue and Zika viruses. Through close
collaboration with the National Public Health Laboratory at the National Centre for Infectious Diseases of the Ministry
of Health we contribute our knowledge and computational expertise at the national frontline for infectious disease
surveillance. With the arrival of a new pandemic through a new Coronavirus causing
COVID-19, the group has once again shown its value in reacting early and
fast in the outbreak to not only help in sharing and analysing genomes
globally via the GISAID platform but also work with other groups to
quickly develop tools for diagnostics, repurposing treatment options and
track mutations of the virus to understand global and local
transmission and monitor phenotypic changes.
We aim at bridging the gap from nucleotide variation to protein structures to interpret effects of human mutations. For example, we have helped clinical collaborators to analyze variants found in patients and tried to mechanistically explain their possible role in a range of diseases like cancer, myopia, leprosy or atopic dermatitis. We are participating in the National Precision Medicine Programme to help mapping mutations into 3D protein structures relative to drug binding sites.
One notable success in the area of human mutation analysis in 2018 was a work led by our Research Scientist Vachiranee.
Using a pathway approach that incorporates comprehensive protein-protein interaction data to systematically extend the
gemcitabine pharmacologic pathway (Figure 1), we identified 77 related nsSNPs, common in the Singaporean population.
Next, we used five computational criteria to prioritize the SNPs based on their importance for protein function. We
specifically selected and screened six candidate SNPs in a patient cohort with non-small cell lung cancer treated with
gemcitabine-based chemotherapy. Using this approach, we were able to identify 3 new SNP biomarkers predictive
for personalized drug treatment outcome while providing a rationale for a molecular mechanism of the SNP effect.
In our ongoing flagship industry project, large multinational Procter & Gamble and BII are jointly developing animaltesting-free Bioinformatics techniques for assessing the allergy potential of proteins using their amino acid sequence and tertiary structure. Often including industry collaborations, we are applying our sequence function and pathway analysis capabilities to support BII’s Natural Product Library and the A*STAR Biotransformation Innovation Platform as well as the Pharma Innovation Programme Singapore. A new direction is to support A*STAR’s Innovations in Food and Chemical Safety programme at the academia-industry interface. In this collaboration with many different groups, our role includes in silico protein allergy prediction for food safety, protein binding target identification, pathway analysis and to highlight common SNPs in the local population that may alter the response to toxic substances.
Sebastian Maurer-Stroh studied theoretical biochemistry at the University of Vienna and wrote his master and PhD thesis at the renowned Institute of Molecular Pathology (IMP). After the honour of a FEBS and a Marie Curie fellowship at the VIB-SWITCH lab in Brussels, he has been leading a group of experts in protein sequence analysis as a senior principal investigator in the A*STAR Bioinformatics Institute (BII) since 2007. He has been appointed Deputy Executive Director (Research) at BII in 2019.
His work includes widely used predictors for posttranslational lipid modifications based on short peptide motifs (4 tools, each cited >100 times), a tool that allows predicting and designing amyloid fibre forming peptides (Nature Methods, cited >300 times since 2010) and he continuously catalyzed new biomolecular insights by sequence-based function predictions (publications in PLoS Genetics, Trends in Biochemical Sciences, Molecular Cell, Current Biology, Genome Biology,…). In 2012, his group also developed the protein sequence search tool TACHYON which is 200 times faster than the all-time classic BLAST. Being able to quickly move from genomes to protein structures through computational analysis and modelling his team is critically contributing to national and global viral pathogen surveillance, most notably with the FluSurver for influenza sequence analysis which has become widely used by National Influenza Centres as part of the global WHO surveillance network and the GISAID initiative. Similar methods can be applied to other viruses and his team critically contributed analyses to recent major virus outbreaks covering H1N1, MERS, H7N9, Ebola, H5N1, Dengue, H5N8/H5N6, Zika and many more.
He also has a strong track record for industry collaborations ranging from local SMEs to large multinationals on sequence analysis and a major research programme on prediction of allergenicity potential of proteins. His protein function analysis skills are also supporting A*STAR's efforts at the academic-private interface through the Biotransformation Innovation Platform (BioTrans), the Pharma Innovation Programme Singapore (PIPS) and the Innovations in Food and Chemical Safety (IFCS) Programme.
Sebastian Maurer-Stroh's research interests lies in mapping the uncharted islands in functional protein sequence space. This includes inferring functions for uncharacterized genes/proteins based on remote evolutionary relationships, prediction of the 3-dimensional structure of proteins, identification of biologically important residues and disease-related mutations, as well as developing predictors for short functional motifs in protein sequences.