Difference between revisions of "Developing analysis pipeline for protein-protein interaction data"

From Master Projects
Jump to: navigation, search
 
Line 2: Line 2:
 
|Contact person=Douwe Molenaar
 
|Contact person=Douwe Molenaar
 
|Contact person2=Rob Haselberg
 
|Contact person2=Rob Haselberg
|Master areas=Systems Biology
 
 
|Fulfilled=No
 
|Fulfilled=No
 
}}
 
}}
 +
 +
|free text=}}
 
The bioactivity of proteins is based on their capacity to bind to their targets, which typically are protein receptors, enzymes or low-molecular weight ligands. Therefore, the availability of efficient analytical tools for the study of protein-protein interactions is essential in protein science. Interestingly, most of these methodologies are not suitable for the simultaneous determination of affinity constants of multiple sample components (such as protein isoforms and/or impurities). Recently, in our lab we developed an approach based on capillary electrophoresis (CE) that enables this simultaneous determination of affinities. During the experiments we produce multiple data sets (replicates information and isoform information). In literature there are at least four different mathematical methods described to calculate affinities from CE data sets. We would need to find an approach that allows us in an unbiased way to compare the different mathematical models in order to make a decision on which one to use. This should include the use of proper statistics for the fit of the models, on replicates, as well as to determine outliers. A part of the work would also be directed in finding out in which format the data should be presented in order to get the most reliable results.
 
The bioactivity of proteins is based on their capacity to bind to their targets, which typically are protein receptors, enzymes or low-molecular weight ligands. Therefore, the availability of efficient analytical tools for the study of protein-protein interactions is essential in protein science. Interestingly, most of these methodologies are not suitable for the simultaneous determination of affinity constants of multiple sample components (such as protein isoforms and/or impurities). Recently, in our lab we developed an approach based on capillary electrophoresis (CE) that enables this simultaneous determination of affinities. During the experiments we produce multiple data sets (replicates information and isoform information). In literature there are at least four different mathematical methods described to calculate affinities from CE data sets. We would need to find an approach that allows us in an unbiased way to compare the different mathematical models in order to make a decision on which one to use. This should include the use of proper statistics for the fit of the models, on replicates, as well as to determine outliers. A part of the work would also be directed in finding out in which format the data should be presented in order to get the most reliable results.
  
 
'''Skills''': Analysis of large data sets (using R, for example), SQL.
 
'''Skills''': Analysis of large data sets (using R, for example), SQL.

Latest revision as of 10:04, 24 August 2016


About Developing analysis pipeline for protein-protein interaction data


Description

|free text=}} The bioactivity of proteins is based on their capacity to bind to their targets, which typically are protein receptors, enzymes or low-molecular weight ligands. Therefore, the availability of efficient analytical tools for the study of protein-protein interactions is essential in protein science. Interestingly, most of these methodologies are not suitable for the simultaneous determination of affinity constants of multiple sample components (such as protein isoforms and/or impurities). Recently, in our lab we developed an approach based on capillary electrophoresis (CE) that enables this simultaneous determination of affinities. During the experiments we produce multiple data sets (replicates information and isoform information). In literature there are at least four different mathematical methods described to calculate affinities from CE data sets. We would need to find an approach that allows us in an unbiased way to compare the different mathematical models in order to make a decision on which one to use. This should include the use of proper statistics for the fit of the models, on replicates, as well as to determine outliers. A part of the work would also be directed in finding out in which format the data should be presented in order to get the most reliable results.

Skills: Analysis of large data sets (using R, for example), SQL.