AD-Genetics-Prediction / Git / Diff of /README.md

Models:

MarcoTheBlack/

AD-Genetics-Prediction

Downloads: 1

Diff of /README.md [4fba4e] .. [61e40d]

Switch to unified view


# AD-Genetics-Prediction
A large-scale phenotype-based AD disease gene prediction

## Introduction
Alzheimer’s disease (AD) is a severe neurodegenerative disorder and has become a global public health problem. Intensive research has been conducted for AD. But the pathophysiology of AD is still not elucidated. Disease comorbidity often associates diseases with overlapping patterns of genetic markers. This may inform a common etiology and suggest essential protein targets. US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) collects large-scale post-marketing surveillance data that provide a unique opportunity to investigate disease co-occurrence pattern. We aim to construct a heterogeneous network that integrates disease comorbidity network from FAERS with protein-protein interaction to prioritize the AD risk genes using network-based ranking algorithm.







## Documentation for modules
1. __*assoc_rules:*__ Provides classes for generate and processing association rules from FAERS using FP-Growth algorithm
1. __*network:*__ Provides classes for DCN and DCN0PPI construction
1. __*graph_algorithm:*__ Provides classes for graph algorithms used in this project, including random walk with restart (rwr),random graph generation and *do novo* predition of AD risk genes
1. __*util:*__ Provides utility classes used in this project

## Results
This folder contains four files listed below.

1. __*fares_comm_net_lift_final_abbr:*__ DCN network file that contains 1,538 disease nodes and 21,312 edges, which was extracted from FAERS using Association Rule Mining
  
    Format: 
    dis1_UMLS|dis1_name|dis1_SOC|dis1_SOC_idx|dis2_UMLS|dis2_name|dis2_SOC|dis2_SOC_idx|conf

    * *dis1_UMLS:* UMLS code for disorder 1
    * *dis1_name:* Name for disorder 1
    * *dis1_SOC:* System Organ Class (MedDRA) for disorder 1
    * *dis1_SOC_idx:* SOC index for disorder 1
    * *dis2_UMLS:* UMLS code for disorder 2
    * *dis2_name:* Name for disorder 2
    * *dis2_SOC:* System Organ Class (MedDRA) for disorder 2
    * *dis2_SOC_idx:* SOC index for disorder 2
    * *conf:* Confidence for disease pair relationship. Since this is an undirected and unweighted graph, value is set to 1.0.

2.  __*DCN_PPI_net.txt:*__ The heterogeneous network file that contains 19,398 nodes (1,538 disease nodes and 17,860 gene nodes) and 1,401,358 edges.

    Format: UMLS_ID or gene symbol|UMLS_ID or gene symbol|weight
    
    Note: The edge weigh is set to 1.0 since the network is undirected and unweighted.

3.  __*disUMLS_name.txt:*__ A disease node mapping file from UMLS ID name to disease concept name

    Format: UMLS_ID|disease_name

4.  __*AD_novel_genes.csv:*__ A file that contains novel AD risk genes we predicted from DCN_PPI network

    Format: Rank,Gene




	a/README.md		b/README.md
1	# AD-Genetics-Prediction	1	# AD-Genetics-Prediction
2	A large-scale phenotype-based AD disease gene prediction	2	A large-scale phenotype-based AD disease gene prediction
3		3
4	## Introduction	4	## Introduction
5	Alzheimer’s disease (AD) is a severe neurodegenerative disorder and has become a global public health problem. Intensive research has been conducted for AD. But the pathophysiology of AD is still not elucidated. Disease comorbidity often associates diseases with overlapping patterns of genetic markers. This may inform a common etiology and suggest essential protein targets. US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) collects large-scale post-marketing surveillance data that provide a unique opportunity to investigate disease co-occurrence pattern. We aim to construct a heterogeneous network that integrates disease comorbidity network from FAERS with protein-protein interaction to prioritize the AD risk genes using network-based ranking algorithm.	5	Alzheimer’s disease (AD) is a severe neurodegenerative disorder and has become a global public health problem. Intensive research has been conducted for AD. But the pathophysiology of AD is still not elucidated. Disease comorbidity often associates diseases with overlapping patterns of genetic markers. This may inform a common etiology and suggest essential protein targets. US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS) collects large-scale post-marketing surveillance data that provide a unique opportunity to investigate disease co-occurrence pattern. We aim to construct a heterogeneous network that integrates disease comorbidity network from FAERS with protein-protein interaction to prioritize the AD risk genes using network-based ranking algorithm.
6		6
7	## Methods	7
8
9	<p align="center">
10	<img src="./figures/methods.jpg" width="400">
11	</p>
12
13	## Documentation for modules	8	## Documentation for modules
14	1. __assoc_rules:__ Provides classes for generate and processing association rules from FAERS using FP-Growth algorithm	9	1. __assoc_rules:__ Provides classes for generate and processing association rules from FAERS using FP-Growth algorithm
15	1. __network:__ Provides classes for DCN and DCN0PPI construction	10	1. __network:__ Provides classes for DCN and DCN0PPI construction
16	1. __graph_algorithm:__ Provides classes for graph algorithms used in this project, including random walk with restart (rwr),random graph generation and do novo predition of AD risk genes	11	1. __graph_algorithm:__ Provides classes for graph algorithms used in this project, including random walk with restart (rwr),random graph generation and do novo predition of AD risk genes
17	1. __util:__ Provides utility classes used in this project	12	1. __util:__ Provides utility classes used in this project
18		13
19	## Results	14	## Results
20	This folder contains four files listed below.	15	This folder contains four files listed below.
21		16
22	1. __fares_comm_net_lift_final_abbr:__ DCN network file that contains 1,538 disease nodes and 21,312 edges, which was extracted from FAERS using Association Rule Mining	17	1. __fares_comm_net_lift_final_abbr:__ DCN network file that contains 1,538 disease nodes and 21,312 edges, which was extracted from FAERS using Association Rule Mining
23		18
24	Format:	19	Format:
25	dis1_UMLS\|dis1_name\|dis1_SOC\|dis1_SOC_idx\|dis2_UMLS\|dis2_name\|dis2_SOC\|dis2_SOC_idx\|conf	20	dis1_UMLS\|dis1_name\|dis1_SOC\|dis1_SOC_idx\|dis2_UMLS\|dis2_name\|dis2_SOC\|dis2_SOC_idx\|conf
26		21
27	* dis1_UMLS: UMLS code for disorder 1	22	* dis1_UMLS: UMLS code for disorder 1
28	* dis1_name: Name for disorder 1	23	* dis1_name: Name for disorder 1
29	* dis1_SOC: System Organ Class (MedDRA) for disorder 1	24	* dis1_SOC: System Organ Class (MedDRA) for disorder 1
30	* dis1_SOC_idx: SOC index for disorder 1	25	* dis1_SOC_idx: SOC index for disorder 1
31	* dis2_UMLS: UMLS code for disorder 2	26	* dis2_UMLS: UMLS code for disorder 2
32	* dis2_name: Name for disorder 2	27	* dis2_name: Name for disorder 2
33	* dis2_SOC: System Organ Class (MedDRA) for disorder 2	28	* dis2_SOC: System Organ Class (MedDRA) for disorder 2
34	* dis2_SOC_idx: SOC index for disorder 2	29	* dis2_SOC_idx: SOC index for disorder 2
35	* conf: Confidence for disease pair relationship. Since this is an undirected and unweighted graph, value is set to 1.0.	30	* conf: Confidence for disease pair relationship. Since this is an undirected and unweighted graph, value is set to 1.0.
36		31
37	2. __DCN_PPI_net.txt:__ The heterogeneous network file that contains 19,398 nodes (1,538 disease nodes and 17,860 gene nodes) and 1,401,358 edges.	32	2. __DCN_PPI_net.txt:__ The heterogeneous network file that contains 19,398 nodes (1,538 disease nodes and 17,860 gene nodes) and 1,401,358 edges.
38		33
39	Format: UMLS_ID or gene symbol\|UMLS_ID or gene symbol\|weight	34	Format: UMLS_ID or gene symbol\|UMLS_ID or gene symbol\|weight
40		35
41	Note: The edge weigh is set to 1.0 since the network is undirected and unweighted.	36	Note: The edge weigh is set to 1.0 since the network is undirected and unweighted.
42		37
43	3. __disUMLS_name.txt:__ A disease node mapping file from UMLS ID name to disease concept name	38	3. __disUMLS_name.txt:__ A disease node mapping file from UMLS ID name to disease concept name
44		39
45	Format: UMLS_ID\|disease_name	40	Format: UMLS_ID\|disease_name
46		41
47	4. __AD_novel_genes.csv:__ A file that contains novel AD risk genes we predicted from DCN_PPI network	42	4. __AD_novel_genes.csv:__ A file that contains novel AD risk genes we predicted from DCN_PPI network
48		43
49	Format: Rank,Gene	44	Format: Rank,Gene
50		45
51		46
52		47