BIN-PPI-Databases
Protein-Protein Interaction Databases
(IntAct, iRef,)
Abstract:
Exploring IntAct and BioGRID PPI databases.
|
Objectives:
|
Outcomes:
|
Deliverables:
Prerequisites:
You need the following preparation before beginning this unit. If you are not familiar with this material from courses you took previously, you need to prepare yourself from other information sources:
- Biomolecules: The molecules of life; nucleic acids and amino acids; the genetic code; protein folding; post-translational modifications and protein biochemistry; membrane proteins; biological function.
This unit builds on material covered in the following prerequisite units:
Evaluation
Evaluation: NA
Contents
In high-throughput biology, the genome was the beginning. As Sydney Brenner has phrased it: we have now written the "white-pages" of the cell, fulfilling the "CAP-criterion" (Comprehensive, Accurate and Permanent). The next level is figuring out the way the parts work - if you will, the "Yellow Pages" - and many of us expect that substantial progress can be made by mapping their interactions. After all, physiological function can be described to a large part as the result of physical interaction.
Please note that there are different types of physical interactions. We most often think of complexes, either stable or transient homo- or heterooligomers when we speak of physical interactions. But there are also interactions between substrates and products and not all of them correspond to classical enzymatic pathways. Phosphorylation and dephosphorylation are processes of key importance in signal transduction and acetylation/deacetylation plays a critical role in regulatory pathways. Here, the substrates are proteins and the interaction with the modifying enzyme is of course a physical interaction.
Genetic interactions on the other hand are another story. Here the word interaction is used in an entirely different sense: it is not synonymous with contact it is synonymous with influence. In fact, most proteins that display genetic interactions would not be expected to interact physically as well. See the FND-PPI-Physical_vs_genetic unit for details.
Task:
- Read the introductory notes on protein-protein interaction databases.
- Read
| Licata & Orchard (2016) The MIntAct Project and Molecular Interaction Databases. Methods Mol Biol 1415:55-69. (pmid: 27115627) |
|
[ PubMed ] [ DOI ] Molecular interaction databases collect, organize, and enable the analysis of the increasing amounts of molecular interaction data being produced and published as we move towards a more complete understanding of the interactomes of key model organisms. The organization of these data in a structured format supports analyses such as the modeling of pairwise relationships between interactors into interaction networks and is a powerful tool for understanding the complex molecular machinery of the cell. This chapter gives an overview of the principal molecular interaction databases, in particular the IMEx databases, and their curation policies, use of standardized data formats and quality control rules. Special attention is given to the MIntAct project, in which IntAct and MINT joined forces to create a single resource to improve curation and software development efforts. This is exemplified as a model for the future of molecular interaction data collation and dissemination. |
| Oughtred et al. (2019) The BioGRID interaction database: 2019 update. Nucleic Acids Res 47:D529-D541. (pmid: 30476227) |
|
[ PubMed ] [ DOI ] The Biological General Repository for Interaction Datasets (BioGRID: https://thebiogrid.org) is an open access database dedicated to the curation and archival storage of protein, genetic and chemical interactions for all major model organism species and humans. As of September 2018 (build 3.4.164), BioGRID contains records for 1 598 688 biological interactions manually annotated from 55 809 publications for 71 species, as classified by an updated set of controlled vocabularies for experimental detection methods. BioGRID also houses records for >700 000 post-translational modification sites. BioGRID now captures chemical interaction data, including chemical-protein interactions for human drug targets drawn from the DrugBank database and manually curated bioactive compounds reported in the literature. A new dedicated aspect of BioGRID annotates genome-wide CRISPR/Cas9-based screens that report gene-phenotype and gene-gene relationships. An extension of the BioGRID resource called the Open Repository for CRISPR Screens (ORCS) database (https://orcs.thebiogrid.org) currently contains over 500 genome-wide screens carried out in human or mouse cell lines. All data in BioGRID is made freely available without restriction, is directly downloadable in standard formats and can be readily incorporated into existing applications via our web service platforms. BioGRID data are also freely distributed through partner model organism databases and meta-databases. |
Data Sources
Interaction databases have similar problems as sequence databases: the need for standards for abstracting biological concepts into computable objects, data integrity, search and retrieval, and the metrics of comparison. There is however an added complication: interactions are rarely all-or-none, and the high-throughput experimental methods have large false-positive and false-negative rates. This makes it necessary to define confidence scores for interactions. On top of experimental methods, there are also a variety of methods for computational interaction prediction. However, even though the "gold standard" are careful, small-scale laboratory experiments, different curated efforts on the same experimental publication usually lead to different results - with as little as 42% overlap between databases being reported.
Currently, likely the best integrated protein-protein interaction database is IntAct, at the EBI, which, besides curating interactions from the literature, hosts interactions from the IMEx consortium = an extensive data-sharing agreement between a number of general and specialized source databases.
Task:
- Access IntAct and enter the UniProt ID for yeast Mbp1 P39678.
- The EBI search directly returns a table of pairwise interactions; both partners are listed as a pair and, in each pair one of the partners should be "Mbp1", or "YDL056W"" (the systematic name of yeast Mbp1).
- How many different physical interaction detection methods do the IntAct records list? Follow the links and read their definitions. (Bravo to the IntAct developers, for defining their terms. In a better world, all the semantics of our databases should be similarly defined to be meaningful.)
But now what?
If you are like me, you would like to be able to link expression profiles, information about known complexes, GO annotations, knock-out phenotypes etc. etc. Not on the Web.
Next, we explore the BioGRID interaction database. BioGrid stores physical and genetic interactions.
Task:
- Access the the BioGRID database at the Samuel-Lunenfeld Research Institute, Mount Sinai Hospital, Toronto. Search for interactions of the Mbp1 gene by entering the gene name into the form field.
- Follow the correct link in BioGrid for saccharomyces cerevisiae Mbp1 (YDL056W). All genes listed in that table have demonstrated interactions with Mbp1.
- List what general experimental type(s) the BioGrid interactors come from. (In particular note the difference between yellow and green boxes).
You will note that some, but not all physical interactions listed by BioGRID and IntAct are the same according to a restrictive interpretation: same organism, same proteins, same experiment, same publication.
- Which of the IntAct Mbp1 interactions are the same in BioGrid?
- Check whether all of the interactions between the regulators of the G1/S phase as per the digram in the "Systems Concepts" PDF are present in BioGRID interactions.
Now, what about MYSPE? Could you infer interactions between proteins whose orthologs interact in another species? Such predictions are called interologs (interacting homologs). Unfortunately, that does not appear to be the case. Confident prediction of interologs can only be achieved in cases of >80% joint sequence identity of both pairs[1], a level of similarity that (I believe) none of our Mbp1 proteins achieves. Does this mean the pathways and interactions are not conserved? Certainly not. We expect a very high degree of conservation of the system's function, but we can't say for sure whether any two specific proteins interact in a different species the same way they interact in yeast. All we can do is to use annotation transfer for hypothesis generation. But that is a useful starting point.
Further reading, links and resources
| Lewis et al. (2012) What evidence is there for the homology of protein-protein interactions?. PLoS Comput Biol 8:e1002645. (pmid: 23028270) |
|
[ PubMed ] [ DOI ] The notion that sequence homology implies functional similarity underlies much of computational biology. In the case of protein-protein interactions, an interaction can be inferred between two proteins on the basis that sequence-similar proteins have been observed to interact. The use of transferred interactions is common, but the legitimacy of such inferred interactions is not clear. Here we investigate transferred interactions and whether data incompleteness explains the lack of evidence found for them. Using definitions of homology associated with functional annotation transfer, we estimate that conservation rates of interactions are low even after taking interactome incompleteness into account. For example, at a blastp E-value threshold of 10(-70), we estimate the conservation rate to be about 11 % between S. cerevisiae and H. sapiens. Our method also produces estimates of interactome sizes (which are similar to those previously proposed). Using our estimates of interaction conservation we estimate the rate at which protein-protein interactions are lost across species. To our knowledge, this is the first such study based on large-scale data. Previous work has suggested that interactions transferred within species are more reliable than interactions transferred across species. By controlling for factors that are specific to within-species interaction prediction, we propose that the transfer of interactions within species might be less reliable than transfers between species. Protein-protein interactions appear to be very rarely conserved unless very high sequence similarity is observed. Consequently, inferred interactions should be used with care. |
| Garcia-Garcia et al. (2012) BIPS: BIANA Interolog Prediction Server. A tool for protein-protein interaction inference. Nucleic Acids Res 40:W147-51. (pmid: 22689642) |
|
[ PubMed ] [ DOI ] Protein-protein interactions (PPIs) play a crucial role in biology, and high-throughput experiments have greatly increased the coverage of known interactions. Still, identification of complete inter- and intraspecies interactomes is far from being complete. Experimental data can be complemented by the prediction of PPIs within an organism or between two organisms based on the known interactions of the orthologous genes of other organisms (interologs). Here, we present the BIANA (Biologic Interactions and Network Analysis) Interolog Prediction Server (BIPS), which offers a web-based interface to facilitate PPI predictions based on interolog information. BIPS benefits from the capabilities of the framework BIANA to integrate the several PPI-related databases. Additional metadata can be used to improve the reliability of the predicted interactions. Sensitivity and specificity of the server have been calculated using known PPIs from different interactomes using a leave-one-out approach. The specificity is between 72 and 98%, whereas sensitivity varies between 1 and 59%, depending on the sequence identity cut-off used to calculate similarities between sequences. BIPS is freely accessible at http://sbi.imim.es/BIPS.php. |
| Keskin et al. (2016) Predicting Protein-Protein Interactions from the Molecular to the Proteome Level. Chem Rev 116:4884-909. (pmid: 27074302) |
|
[ PubMed ] [ DOI ] Identification of protein-protein interactions (PPIs) is at the center of molecular biology considering the unquestionable role of proteins in cells. Combinatorial interactions result in a repertoire of multiple functions; hence, knowledge of PPI and binding regions naturally serve to functional proteomics and drug discovery. Given experimental limitations to find all interactions in a proteome, computational prediction/modeling of protein interactions is a prerequisite to proceed on the way to complete interactions at the proteome level. This review aims to provide a background on PPIs and their types. Computational methods for PPI predictions can use a variety of biological data including sequence-, evolution-, expression-, and structure-based data. Physical and statistical modeling are commonly used to integrate these data and infer PPI predictions. We review and list the state-of-the-art methods, servers, databases, and tools for protein-protein interaction prediction. |
Notes
- ↑
Mika & Rost (2006) Protein-protein interactions more conserved within species than across species. PLoS Comput Biol 2:e79. (pmid: 16854211) [ PubMed ] [ DOI ] Experimental high-throughput studies of protein-protein interactions are beginning to provide enough data for comprehensive computational studies. Today, about ten large data sets, each with thousands of interacting pairs, coarsely sample the interactions in fly, human, worm, and yeast. Another about 55,000 pairs of interacting proteins have been identified by more careful, detailed biochemical experiments. Most interactions are experimentally observed in prokaryotes and simple eukaryotes; very few interactions are observed in higher eukaryotes such as mammals. It is commonly assumed that pathways in mammals can be inferred through homology to model organisms, e.g. the experimental observation that two yeast proteins interact is transferred to infer that the two corresponding proteins in human also interact. Two pairs for which the interaction is conserved are often described as interologs. The goal of this investigation was a large-scale comprehensive analysis of such inferences, i.e. of the evolutionary conservation of interologs. Here, we introduced a novel score for measuring the overlap between protein-protein interaction data sets. This measure appeared to reflect the overall quality of the data and was the basis for our two surprising results from our large-scale analysis. Firstly, homology-based inferences of physical protein-protein interactions appeared far less successful than expected. In fact, such inferences were accurate only for extremely high levels of sequence similarity. Secondly, and most surprisingly, the identification of interacting partners through sequence similarity was significantly more reliable for protein pairs within the same organism than for pairs between species. Our analysis underlined that the discrepancies between different datasets are large, even when using the same type of experiment on the same organism. This reality considerably constrains the power of homology-based transfer of interactions. In particular, the experimental probing of interactions in distant model organisms has to be undertaken with some caution. More comprehensive images of protein-protein networks will require the combination of many high-throughput methods, including in silico inferences and predictions. http://www.rostlab.org/results/2006/ppi_homology/
About ...
Author:
- Boris Steipe <boris.steipe@utoronto.ca>
Created:
- 2017-08-05
Modified:
- 2020-09-22
Version:
- 1.1
Version history:
- 1.1 2020 Update
- 1.0 First live
- 0.1 First stub
This copyrighted material is licensed under a Creative Commons Attribution 4.0 International License. Follow the link to learn more.
