Introduction
Background and literature review
Research design and methodology
Data resources
Data features
Drug
Gene
Disease
Prediction model
Side effect prediction
ATC | The number of side effect | ATC | The number of side effect |
---|---|---|---|
ALL | 2326 | L | 970 |
A | 642 | M | 652 |
B | 264 | N | 1704 |
C | 632 | P | 113 |
D | 379 | R | 380 |
G | 523 | S | 914 |
H | 132 | V | 191 |
J | 702 |
Incidence prediction
Dataset No. | SCORE | ATC | Gene ID | Incidence | Items |
---|---|---|---|---|---|
1 | ∘ | ∘ | ∘ | 358 | |
2 | ∘ | ∘ | 262 | ||
3 | ∘ | ∘ | 214 | ||
4 | ∘ | ∘ | ∘ | ∘ | 469 |
5 | ∘ | ∘ | ∘ | 334 | |
6 | ∘ | ∘ | ∘ | 469 | |
7 | ∘ | ∘ | 334 |
Dataset No. | Clustering | Correct rate (%) | Dataset No. | Clustering | Correct rate (%) |
---|---|---|---|---|---|
1 | 41 –100 % | 99 | 5 | 34.7 –100 % | 100 |
2 | 55 –100 % | 96 | 27.4 –100 % | 100 | |
29 –100 % | 100 | 12.1 –100 % | 100 | ||
9 –100 % | 100 | 6 | 53.2 –100 % | 95 | |
3 | 70.3 –100 % | 100 | 51 –100 % | 100 | |
36 –100 % | 99 | 50 –100 % | 100 | ||
18 –100 % | 100 | 34.7 –100 % | 100 | ||
4 | 81.7 –100 % | 100 | 27.4 –100 % | 100 | |
63.5 –100 % | 100 | 18 –100 % | 100 | ||
34.7 –100 % | 100 | 12.1 –100 % | 100 | ||
27.4 –100 % | 100 | 7 | 81.7 –100 % | 100 | |
12.1 –100 % | 100 | 49 –100 % | 100 | ||
5 | 94 –100 % | 100 | 18 –100 % | 100 |
Dataset No. | Clustering | Correct rate (%) | Dataset No. | Clustering | Correct rate (%) |
---|---|---|---|---|---|
1 | 41 –100 % | 100 | 5 | 34.7 –100 % | 100 |
0.1 –100 % | 100 | 27.4 –100 % | 99.5 | ||
2 | 55 –100 % | 23 | 12.1 –100 % | 99.6 | |
29 –100 % | 95 | 0.1 –100 % | 100 | ||
9 –100 % | 40 | 6 | 53.2 –100 % | 100 | |
0.1 –100 % | 100 | 51 –100 % | 100 | ||
3 | 70.3 –100 % | 100 | 50 –100 % | 100 | |
36 –100 % | 90 | 34.7 –100 % | 100 | ||
18 –100 % | 20 | 27.4 –100 % | 98 | ||
0.1 –100 % | 86 | 18 –100 % | 98 | ||
4 | 81.7 –100 % | 100 | 12.1 –100 % | 99 | |
63.5 –100 % | 100 | 0.1 –100 % | 100 | ||
34.7 –100 % | 100 | 7 | 81.7 –100 % | 100 | |
27.4 –100 % | 100 | 49 –100 % | 100 | ||
12.1 –100 % | 100 | 18 –100 % | 99 | ||
0.1 –100 % | 100 | 0.1 –100 % | 100 | ||
5 | 94 –100 % | 100 |
Data modeling
Domain model
Logical system model
Implementation model
Implementation
System overview
Software overview
Open Source |
OS: Linux Fedora 20 |
IDE: Eclipse 4.3.2 (KEPLER) |
Java: 1.7.0_55 |
Application Framework: Struts 1.3.10 |
Web container/HTTP/Web server: Tomcat 7.0.53 |
Database: HSQLDB 2.3.2 |
Machine learning library: LibSVM 3.18 |
Free ware |
Domain object: Book Management System (bcat) |
Application
Results and discussion
Side effect prediction
Dataset No. | Approximated interval | Higher interval |
---|---|---|
1 | 99.7 % | 0.3 % |
2 | 97 % | 3 % |
3 | 11 % | 89 % |
4 | 98 % | 2 % |
5 | 76 % | 24 % |
6 | 71 % | 29 % |
7 | 44 % | 56 % |