Identify protein interactions that may be impacted by splicing of PTMs

Post translational modifications (PTMs) often facilitate protein interactions, either through direct binding of domains specific to that particular modification (e.g. SH2 domains binding to phosphorylated tyrosines) or through allosteric effects that change the conformation of the protein to either enhance or disrupt interactions. We provide functions to annotate spliced PTMs with relevant protein interactions and to identify key PTMs that may disrupt protein interaction networks.

Currently, we provide functions to process and analyze protein interaction data from PhosphoSitePlus, PTMInt, and PTMcode. We can also include enzyme-specific interactions (such as kinase substrate interactions through PhosphoSitePlus and RegPhos). First, we need to annotate the spliced PTMs with protein interactions (see rest of documentation for how to do this). Then, we can process the interactions across the different databases using the protein_interactions class to identify key PTMs that may disrupt protein interaction networks.

from ptm_pose import analyze
import pandas as pd

# Load spliced ptm and altered flank data
spliced_ptms = pd.read_csv('spliced_ptms.csv')

interactions = analyze.protein_interactions(spliced_ptms)
interactions.get_interaction_network()

interactions.network_data.head()

PhosphoSitePlus regulatory site data found and added
Combined kinase-substrate data found and added
PTMcode data found and added
PTMInt data found and added
ELM data found and added

	Modified Gene	Interacting Gene	Residue	Type	Source	dPSI	Regulation Change
0	ABI1	ABL1	S361	INDUCES	PTMcode	0.213	+
1	ABI1	BAIAP2	S361	INDUCES	PTMcode	0.213	+
2	ABI1	CYFIP2	S361	INDUCES	PTMcode	0.213	+
3	ABI1	EPS8	S361	INDUCES	PTMcode	0.213	+
4	ABI1	EPS8L1	S361	INDUCES	PTMcode	0.213	+

We can also calculate interaction stats to identify proteins that are most impacted or relevant to spliced PTMs and the protein interaction network

interactions.get_interaction_stats()

interactions.network_stats.head()

	Degree	Degree Centrality	Closeness	Betweenness
ABI1	18	0.111111	0.214493	0.237466
ABL1	3	0.018519	0.176786	0.105207
BAIAP2	2	0.012346	0.166163	0.000000
CYFIP2	1	0.006173	0.164051	0.000000
EPS8	1	0.006173	0.164051	0.000000

If we want to focus on a specific protein, we can summarize information about a single protein in the network. In this case, let’s look at TSC2, which loses pS981 upon ESRP1 knockdown

interactions.summarize_protein_network(protein = 'TSC2')

Decreased interaction likelihoods: AKT1, SGK1, YWHAE, YWHAZ
Number of interactions: 4 (Rank: 14)
Centrality measures -    Degree = 0.024691358024691357 (Rank: 14)
                         Betweenness = 0.0004600874166091557 (Rank: 28)
                         Closeness = 0.024691358024691357 (Rank: 126)

We can also visualize the network…

interactions.plot_interaction_network(interacting_node_size = 10)

…and the centrality of proteins in the network

interactions.plot_network_centrality(centrality_measure='Degree')