Week 3 - TOOL - Identifying Protein-Protein Interactions Using the String Bioinformatics Tool

Proteins do not exist in isolation within a cell. To perform their biological roles, they frequently interact with other proteins and molecules.
These interactions are essential for executing functions required to maintain complex cellular processes.
String is a specific bioinformatics tool designed to help researchers identify these protein-protein interactions (PPIs).

Users identify a protein of interest by typing its name into the search tool; for example, searching for "tuberin."
A specific organism must be selected to narrow the results. In the case of studying human cellular biology, the organism selected is Homo sapiens.
When multiple results appear, the most relevant entry is typically at the top of the list.

The primary output of a String search is a visual network graph representing protein-protein interactions.
The graph shows how a specific protein (such as tuberin, which is encoded by the gene TSC2) interacts with various other identified proteins.
Many of the proteins identified in the results also interact with one another, forming an interconnected web rather than just a simple list of partners for the target protein.
The visual interface allows for manual manipulation; users can click and pull on the protein nodes to spread the network out, making it easier to see individual connections.

Every line drawn between two protein nodes in the network represents existing evidence suggesting those proteins interact.
An alternative way to interpret these findings is through the Predicted Functional Partners table located at the bottom of the tool's interface.
Specific pieces of evidence are categorized into various types, including:
- Text mining
- Databases
- Experiments
- Co-expression
A visual indicator for the strength of evidence is provided via dots; the darker the dot, the stronger the evidence for that specific interaction.
Quantitative Likelihood Scores:
- Interactions are assigned a likelihood score out of a maximum value of $1$ .
- Scores very close to $1$ indicate a very high likelihood that the predicted interaction is a real occurrence within the cell.

The relationship between TSC2 and TSC1 serves as a prime example of high-confidence interaction.
The protein product of the TSC1 gene is named hamartin, and the protein product of the TSC2 gene is named tuberin.
The tool displays $4$ distinct lines of evidence suggesting their interaction, confirming that these two proteins form a dimer together.
In certain visual representations, a protein might show three dark dots and one lighter dot across different evidence categories, supported by a high numerical score.

String provides succinct descriptions of the biological functions of individual proteins.
Accessing these descriptions is done by either hovering the cursor over a protein node or clicking on it directly.
Function of Tuberin (TSC2):
- It is identified as a TIM suppressor.
- It is involved in the inhibition of mTOR.
Function of mTOR:
- mTOR is a critical transcription factor.
- It is responsible for the up-regulation of the cell cycle.
Function of Reb:
- Reb is identified as a GTP binding protein.
- It possesses kinase activity.
- It is involved in the activation of the mTOR transcription factor.
Structural Information: If a protein's structure has been determined, the tool may display it alongside the functional description.

Understanding protein-protein interactions is vital for determining the specific role a protein performs and how it contributes to cellular function.
Researchers use String to examine if there is any functional crossover between proteins.
By analyzing the functions of multiple interacting proteins, it is possible to determine if they work together within a single cellular process or across multiple different processes.
Identifying these interactions is a critical step when building resources like protein-based disease websites to explain why specific interactions are important for a protein of interest.