Predicting synthetic lethal interactions using conserved patterns in protein interaction networks

Benstead-Hume, Graeme; Chen, Xiangrong; Hopkins, Suzanna R.; Lane, Karen A.; Downs, Jessica A.; Pearl, Frances M. G.

doi:10.1371/journal.pcbi.1006888

Cited by 35 publications

(38 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Analysis of these and other network-based features has been used successfully in the prediction of essential genes across diverse organisms [64]. Prediction of synthetic lethal relationships between genes is another potential network use, for instance Benstead-Hume and colleagues used proteinprotein interaction networks to predict human synthetic lethal interactions, which they then confirmed experimentally [65]. Such analyses may be used to identify targets for new chemotherapies in helminth research [66].…”

Section: Discussionmentioning

confidence: 99%

The tapeworm interactome: inferring confidence scored protein-protein interactions from the proteome of Hymenolepis microstoma

James

Olson

2020

BMC Genomics

View full text Add to dashboard Cite

Background: Reference genome and transcriptome assemblies of helminths have reached a level of completion whereby secondary analyses that rely on accurate gene estimation or syntenic relationships can be now conducted with a high level of confidence. Recent public release of the v.3 assembly of the mouse bile-duct tapeworm, Hymenolepis microstoma, provides chromosome-level characterisation of the genome and a stabilised set of protein coding gene models underpinned by bioinformatic and empirical data. However, interactome data have not been produced. Conserved protein-protein interactions in other organisms, termed interologs, can be used to transfer interactions between species, allowing systems-level analysis in non-model organisms. Results: Here, we describe a probabilistic, integrated network of interologs for the H. microstoma proteome, based on conserved protein interactions found in eukaryote model species. Almost a third of the 10,139 gene models in the v.3 assembly could be assigned interaction data and assessment of the resulting network indicates that topologically-important proteins are related to essential cellular pathways, and that the network clusters into biologically meaningful components. Moreover, network parameters are similar to those of single-species interaction networks that we constructed in the same way for S. cerevisiae, C. elegans and H. sapiens, demonstrating that information-rich, system-level analyses can be conducted even on species separated by a large phylogenetic distance from the major model organisms from which most protein interaction evidence is based. Using the interolog network, we then focused on sub-networks of interactions assigned to discrete suites of genes of interest, including signalling components and transcription factors, germline multipotency genes, and genes differentially-expressed between larval and adult worms. Results show not only an expected bias toward highly-conserved proteins, such as components of intracellular signal transduction, but in some cases predicted interactions with transcription factors that aid in identifying their target genes. Conclusions: With key helminth genomes now complete, systems-level analyses can provide an important predictive framework to guide basic and applied research on helminths and will become increasingly informative as new protein-protein interaction data accumulate.

show abstract

Section: Discussionmentioning

confidence: 99%

The tapeworm interactome: inferring confidence scored protein-protein interactions from the proteome of Hymenolepis microstoma

James

Olson

2020

BMC Genomics

View full text Add to dashboard Cite

show abstract

“…Analysis of these and other network-based features has been used successfully in the prediction of essential genes across diverse organisms [Azhagesan et al, 2018]. Prediction of synthetic lethal relationships between genes is another potential network use, for instance Benstead-Hume and colleagues used protein-protein interaction networks to predict human synthetic lethal interactions, which they then confirmed experimentally [Benstead-Hume et al, 2019]. Such analyses could potentially be used to identify targets for new chemotherapies in helminth research [Pinto et al, 2014].…”

Section: Discussionmentioning

confidence: 99%

The tapeworm interactome: inferring confidence scored protein-protein interactions from the proteome of Hymenolepis microstoma

James

Olson

2019

Preprint

View full text Add to dashboard Cite

AbstractReference genome and transcriptome assemblies of helminths have reached a level of completion whereby secondary analyses that rely on accurate gene estimation or syntenic relationships can be now conducted with a high level of confidence. Recent public release of the v.3 assembly of the mouse bile-duct tapeworm, Hymenolepis microstoma, provides chromosome-level characterisation of the genome and a stabilised set of protein coding gene models underpinned by both bioinformatic and empirical data. However, interactome data have not been produced. Conserved protein-protein interactions in other organisms, termed interologs, can be used to transfer interactions between species, allowing systems-level analysis in non-model organisms. Here, we describe a probabilistic, integrated network of interologs for the H. microstoma proteome, based on conserved protein interactions found in eukaryote model species. Almost a third of the 10,139 gene models in the v.3 assembly could be assigned interaction data and assessment of the resulting network indicates that topologically-important proteins are related to essential cellular pathways, and that the network clusters into biologically meaningful components. Moreover, network parameters are similar to those of single-species interaction networks that we constructed in the same way for S. cerevisiae, C. elegans and H. sapiens, demonstrating that information-rich, system-level analyses can be conducted even on species separated by a large phylogenetic distance from the major model organisms from which most protein interaction evidence is based. Using the interolog network, we then focused on sub-networks of interactions assigned to discrete suites of genes of interest, including signalling components and transcription factors, germline ‘multipotency’ genes, and differentially-expressed genes between larval and adult worms. These analyses not only showed an expected bias toward highly-conserved proteins, such as components of intracellular signal transduction, but in some cases predicted interactions with transcription factors that aid in identifying their target genes. With the completion of key helminth genomes, such systems level analyses can provide an important predictive framework to guide basic and applied research on helminths and will become increasingly informative as protein-protein interaction data accumulate.

show abstract

“…Broadly they can be divided into machine learning and statistical methods. Machine learning methods, including those based on network analytics, rely on labelled data to predict SL pairs, e.g., [22,6,26,10]. However, these methods face the challenge of scarce labels since very few human SL pairs are experimentally confirmed.…”

Section: Introductionmentioning

confidence: 99%

ASTER: A Method to Predict Clinically Actionable Synthetic Lethal Genetic Interactions

Liany

Jeyasekharan

Rajan

2020

Preprint

View full text Add to dashboard Cite

A Synthetic Lethal (SL) interaction is a functional relationship between two genes or functional entities where the loss of either entity is viable but the loss of both is lethal. Such pairs can be used to develop targeted anticancer therapies with fewer side effects and reduced overtreatment. However, finding clinically actionable SL interactions remains challenging. Leveraging large-scale unified gene expression data of both disease-free and cancerous data, we design a new technique, based on statistical hypothesis testing, called ASTER (Analysis of Synthetic lethality by comparison with Tissue-specific disease-free gEnomic and tRanscriptomic data) to identify SL pairs. For large-scale multiple hypothesis testing, we develop an extension called ASTER++ that can utilize additional input gene features within the hypothesis testing framework. Our extensive experiments demonstrate the efficacy of ASTER in accurately identifying SL pairs that are therapeutically actionable in stomach and breast cancers.

show abstract

Predicting synthetic lethal interactions using conserved patterns in protein interaction networks

Cited by 35 publications

References 63 publications

The tapeworm interactome: inferring confidence scored protein-protein interactions from the proteome of Hymenolepis microstoma

The tapeworm interactome: inferring confidence scored protein-protein interactions from the proteome of Hymenolepis microstoma

The tapeworm interactome: inferring confidence scored protein-protein interactions from the proteome of Hymenolepis microstoma

ASTER: A Method to Predict Clinically Actionable Synthetic Lethal Genetic Interactions

Contact Info

Product

Resources

About