BIN-SEQA-Collaboration
Analysis of Collaborating Sequences
(Analysis of collaborating sequences (with R examples); Coexpression; Pathways; Enrichment.)
Abstract:
This unit discusses "collaborating" genes in two examples: from the pathway perspective in the the KEGG pathway database and from the view of co-expression, as exemplified in CoexpressDB.
|
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:
Contents
Evaluation
Evaluation: NA
Contents
Pathways
Pathways are perhaps the earliest biochemical representation of systems of collaborating genes. They are a particularly active area of bioinformatics research, stimulated by the vision of automatically grouping biological entities into meaningful systems, based on their known properties, or the properties of homologues. For instance if Enzyme A produces metabolite 1 and Enzyme B consumes metabolite 1, even a computer can figure out that A and B can be functionally connected. If we stitch all such connections together, we arrive at a description of material flow inside the cell, or of regulatory connections, or of signaling events. Of course, A and B have to be in the same compartment, and metabolite 1 shouldn't be ATP. Or H2O. And if we infer relationships from well-studied model organisms, the components we compare really have to be orthologues. We are not sure yet to what degree automation will be possible. Careful, manual curation of pathway data is going to be with us for some time to come.
KEGG
The Kyoto Encyclopedia of Genes and Genomes is one of the oldest and best curated databases of metabolic and functional pathways. It stores hand-curated pathways for a number of model organisms and supports computational inference for other organisms by determining orthologues.
Task:
- Access the KEGG Web site.
Kegg identifies organisms in the database according to three or four letter codes, eg. homo sapiens → hsa, saccharomyces cerevisiae → sce. Some of our fungi have manually curated genes in annotated in KEGG, however not all.
- Navigate to the KEGG Organisms selection page and find whether MYSPE has been annotated in KEGG by searching for the species name on the page. Note the KEGG three letter species code if it is there. If you are certain it is not there, note that you have not found it, note the full taxonomic lineage of MYSPE (from NCBI Taxonomy DB if it is not in your notes). Then proceed through this task using Cryptococcus neoformans (cne).
- Explore
- Navigate to the KEGG2 entry page, which contains a number of options to search the database contents. I have instantiated the links with searches relevant to yeast Mbp1 and the cell-cycle in the list below, try the links and make sure you understand what they contain.
- The simplest option is to search for a gene name. KEGG will return all matches to that name in record titles and annotations.
- You can execute a BLAST search against the database and thus search with domain sequences, such as the APSES domain, rather than with entire genes;
- define a ligand or enzyme;
- use the PATHWAY search tool to retrieve information on a particular system.
The gene-search results return a list of genes, one of them should be sce:YDL056W the KEGG code for Mbp1's systematic name. Access that record and click on the Help button on top of the record to find information about what the returned results contain.
Not all of KEGG's curated genes contain a link to a pathway record. However Mbp1 does. There is a line labeled Pathway with a link to the protein's curated pathway information: sce04111[1].
- Follow the link to the yeast cell-cycle pathway.
The position of Mbp1 is emphasized with a red box. Almost all boxes in this reference pathway are green, this indicates that a gene for that component of the pathway has been curated and stored in the database. The boxes are linked to the respective KEGG gene pages. The phases of the cell cycle G1 - S - G2 - M are indicated at the bottom of the chart.
- Use the drop-down menu to switch to the comparative pathway map curated for MYSPE.
Note that this displays the same map, but now some of the boxes are white (the KEGG curators have not annotated an orthologue for these genes in the KO (KEGG Ontology) database) and the green boxes are now linked to your organism's gene instead of yeast. This is a very convenient way to check which components of the well described yeast pathways have been curated as conserved in your organism.
If you explore the various organisms for which this map has been transferred by homology (options menu at the top), you will notice that most organisms have very much fewer genes mapped to this yeast pathway by the KEGG curators (the situation is somewhat better with metabolic pathways, by the way, since computational inference through orthology is less ambiguous). But orthologues are quite easy to find in KEGG too:
Task:
- Return to the yeast version of the cell-cycle pathway map
- Click on the Mbp1 link on the yeast pathway map to take you to the KEGG gene record for the Mbp1 protein. In the SSDB row, click the Ortholog button. Orthologs for all of the organisms (or at least their close relatives in the KEGG database) have been precomputed. It should take only a moment to check that the orthologue in your organism is listed too - even if the box was "white" on the pathway page. This is not an error - it just reflects different levels of annotation, curation and inference.
Once again, we are back at a familiar problem: much and increasingly more of our annotations are based on analogy and inference. We study one system experimentally in a model organism, then we attempt to map the components to another organism. But pursuing the idea of orthology in order to map function is tricky. Even orthologues may have diverged in evolution to distinct and dissimilar functional systems. Note for example that in yeast Mbp1 binds to Swi6 (the MBF complex) and Swi6 can also bind to Swi4, an Mbp1 homologue (the SBF complex). In many CRMs (cis-regulatory modules) the respective binding sites of Mbp1 and Swi4 are closely juxtaposed. However only the Saccharomycotina seem to posess orthologues to Swi4, at least as far as they are more similar to Swi4 than to Mbp1. Phylogenetic analysis finds an Mbp1 paralogue in the fungal cenancestor, which then was at the root of the Swi4/MbpA genes of our tree. Has Mbp1 taken on functions that are distinct from Swi4? MBF and SBF appear to be two complementary systems, presumably each having taken over some part of the space of functions from the other and probably acquired a few novel functions along the way. But the situation in the other fungi cannot be unambiguously inferred from the evidence we have considered.
I hope that this short discussion has illuminated the problems associated with mapping functions between species, based on gene similarity. To paraphrase the issue one more time: we are mapping concepts to biology, but "concepts" and "biology" exist in two different worlds. It is helpful, indeed crucial to explain biology in terms of higher-order concepts. This is what we ultimately mean by "understanding" and indeed, if we would not try this, we would be merely "butterfly-collecting". But never, never fall into the trap of basing your biological conclusions - eg. functional equivalence of biological objects - mechanically on a computed similarity of concepts (such as gene similarity, pathway position, GO annotation etc.). The mapping of concept to object may be arbitrarily imprecise and as a consequence, so is the equivalence, once we apply it to the "real" world.
Co-Expression
Task:
CoExpressdb is a well curated database of pre-calculated co-expression profiles for model organisms. Expression values across a large number of published experiments on the same platform are compared via their coefficient of correlation. Highly correlated genes are either co-regulated, or one gene influences the expression level of the other.
- Navigate to CoExpressdb.
- Enter
Mbp1as a "gene alias" in the search field. - Click on the link to the coexpressed gene list. Do any of the "known" target genes appear here? How do you interpret this result?
Unfortunately, the support for yeast genes is very limited. CoexDB is however an excellent resource to study higher eukaryotic, especially human genes.Refer to the YouTube tutorials for details.
Further reading, links and resources
| Okamura et al. (2015) COXPRESdb in 2015: coexpression database for animal species by DNA-microarray and RNAseq-based expression data with multiple quality assessment systems. Nucleic Acids Res 43:D82-6. (pmid: 25392420) |
|
[ PubMed ] [ DOI ] The COXPRESdb (http://coxpresdb.jp) provides gene coexpression relationships for animal species. Here, we report the updates of the database, mainly focusing on the following two points. For the first point, we added RNAseq-based gene coexpression data for three species (human, mouse and fly), and largely increased the number of microarray experiments to nine species. The increase of the number of expression data with multiple platforms could enhance the reliability of coexpression data. For the second point, we refined the data assessment procedures, for each coexpressed gene list and for the total performance of a platform. The assessment of coexpressed gene list now uses more reasonable P-values derived from platform-specific null distribution. These developments greatly reduced pseudo-predictions for directly associated genes, thus expanding the reliability of coexpression data to design new experiments and to discuss experimental results. |
Notes
- ↑ The cell-cycle pathway
04111should also be discoverable as one of the pathways returned in a search for "cyclin-dependent kinase" as an enzyme, or as one of the pathways returned in a pathway search for "cell-cycle".
About ...
Author:
- Boris Steipe <boris.steipe@utoronto.ca>
Created:
- 2017-08-05
Modified:
- 2020-09-24
Version:
- 1.1
Version history:
- 1.1 2020 Maintenance and updates
- 0.1 First stub
This copyrighted material is licensed under a Creative Commons Attribution 4.0 International License. Follow the link to learn more.
