Our approach consisted of two steps. The first step was to construct a heterogeneous network by using input data including different kinds of nodes and edges. There were three types of nodes in the proposed heterogeneous network: targets, drugs and diseases. There were six different kinds of edges with represented one type of similarity or association: target similarity, drug similarity, disease similarity, known drug–target interaction, known drug-disease association, and known disease-target interaction. Then, at the second step, we tried to predict important links which were not identified in the input data by using of Heter-LP. As we used the Heter-LP for the second step, we had to arrange our network (step 1) according to its input structures, which consisted of six separate matrices: (1) drug similarities, (2) disease similarities, (3) targets similarities, (4) drug-target relations, (5) disease-target relations and (6) drug-disease relations [3].
We tried to find a specialized dataset for mastitis to integrate with the aforementioned data. Unfortunately, no dataset is publically available. It seems that all available data and information related to animal diseases, genes targets, and drugs are only embedding in the publication and there are no comprehensive datasets or repositories for them. Lack of access to this data however did not negatively impact the current analysis because of the generality of the input datasets due to the similarity of this disease in humans and other animals. To specialize the results for E. coli mastitis we added three parts of information to our generated datasets (cases 2 to 4 in the following).
Based on the above descriptions, data sets that were used to construct the input data network for Heter-LP algorithm included:
-
Previously known drugs information gathered by public repositories.
-
Key genes that had a robust bio-signature in response to mastitis specially in E. coli infection. These genes are the results of our recent research findings based on meta-analyses to detect up/down regulated genes by utilizing machine learning algorithms [1] and network analysis [2, 28],
-
Functionally related diseases or biological processes related to bovine mastitis illustrated by Pathway Studio web tool.
-
Relevant drugs and antibiotics to E. coli mastitis gathered by literature analysis.
Finally, by the integration of these data a heterogeneous network is constructed and used as input for the Heter-LP algorithm.
A brief description and the data preparation methods of each part are presented in next subsections (Fig. 1). Different parts of this workflow were described in its caption and in more details in sub-Sect. 2.1 to 2.5.
2.1 Previously known drugs information
As mentioned before six different kinds of data including (1) Drug similarity, (2) Disease similarity, (3) Target similarity, (4) Known drug-disease associations, (5) Known drug–target interactions, and (6) Known disease-target associations, were necessary as input data to develop the network for Heter-LP. The base of these data was gathered previously [3], from some important databases. The data resources are summarized in the Table 1. In order to have an updated version of these data, the last version of data resources to generate the six matrices are provided according to the methods described at GitHub (https://github.com/MLotfiSH/Heter-LP) and DKR site (http://dkr.iut.ac.ir/projects) and in detail in our previously published research [3].
Table 1
Resources of data related to each sub-network and the number of nodes in each one.
Sub-network | Using criterion | Resource | Number of nodes |
In each resource | In total |
Drugs | Chemical substructures similarities | PubChem1 | 1103 | 5089 |
Side effect similarities | SIDER2 | 888 |
Anatomical Therapeutic Chemical (ATC) code similarities | KEGG3 | 4867 |
Diseases | Disease genes similarities | DisGeNET4 | 3295 | 9886 |
The similarity based on ICD-10 classification5 | KEGG | 1366 |
Semantic similarity based on Disease Ontology (DO)6 | DOSE7 package in R | 6560 |
Semantic similarities based on GO8 | GOSemSim9 package in R | 1550 |
Targets | Semantic similarities based on HPO10 | HPOSim11 package in R | 979 | 2940 |
Semantic similarities based on DO | DOSE package in R | 1092 |
Similarities based on KEGG | KEGG | 1132 |
Drug-disease | | Therapeutic Target Database (TTD)12 | Drugs: 6931 | Drugs: 7382 |
Diseases: 1418 |
KEGG | Drugs: 1052 | Diseases: 1970 |
Diseases: 592 |
Drug-target | | DrugBank13 | Drugs:1521 | Drugs:3350 |
Targets:1346 |
KEGG | Drugs:2440 | Targets:1415 |
Targets:335 |
Disease-target | | DiGeNet | Diseases:577 | Diseases:1838 |
Targets:2403 |
KEGG | Diseases:1271 | Targets:4066 |
Targets:2563 |
1 https://pubchem.ncbi.nlm.nih.gov/score_matrix/score_matrix.cgi
2 http://sideeffects.embl.de/
3 Kyoto Encyclopedia of Genes and Genomes ((http://www.kegg.jp)
4 http://www.disgenet.org
5 International Statistical Classification of Diseases and Related Health Problems-10
6 http://disease-ontology.org/
7 Disease Ontology Semantic and Enrichment analysis (https://bioconductor.org/packages/release/bioc/html/DOSE.html)
8 Gene Ontology (http://www.geneontology.org/)
9 https://bioconductor.org/packages/release/bioc/html/GOSemSim.html
10 Human Phenotype Ontology, https://hpo.jax.org/ (http://human-phenotype-ontology.github.io/)
11 https://mran.microsoft.com/snapshot/2014-10-20/web/packages/HPOSim/index.html
12 http://bidd.nus.edu.sg/group/cjttd/
13 http://drugbank.ca
2.2 Disease genes
Previously we identify differentially expressed genes in response to E. coli mastitis [1, 2, 28]. These genes have a robust bio-signature and thereby may be useful biomarker or therapeutic target candidates in mastitis [1, 2]. These genes/proteins are listed in Table 2 and added to the disease-gene relation part of the dataset shown in Table 1.
Table 2
The key genes or regulators with robust bio-signature in response to E. coli mastitis reported to our previous meta-analysis based microarray studies [1, 2]
Row | Gene symbol | Functional group | Gene name (alias) |
1 | CXCL2 | Ligand | Chemokine (C-X-C motif) ligand 2 (GRO3) |
2 | CXCL8 | Ligand | C-X-C motif chemokine ligand 8 (IL-8, IL8) |
3 | GRO1 | Ligand | Chemokine (C-X-C motif) ligand 1 (CXCL1, MGSA) |
4 | CFB | | Complement factor B (BF) |
5 | ZC3H12A | | Zinc finger CCCH-type containing 12A |
6 | CCL20 | Ligand | C-C motif chemokine ligand 20 |
7 | NFKBIZ | | NFKB inhibitor zeta (MAIL) |
8 | S100A9 | Ligand | S100 calcium binding protein A9 |
9 | S100A8 | Ligand | S100 calcium binding protein A8 |
10 | PDE4B | | Phosphodiesterase 4B |
11 | CASP4 | | Caspase 4, apoptosis-related cysteine peptidase (CASP13) |
12 | HP | Ligand | Haptoglobin |
13 | MAPK1 | Protein kinase | Mitogen-activated protein kinase |
14 | TP53 (p53) | Transcription factor | Tumor protein p53 |
15 | SP1 | Transcription factor | Sp1 transcription factor |
16 | MAPK14 | Protein kinase | Mitogen-activated protein kinase 14 |
17 | INS | Ligand | Insulin |
18 | EGF | Ligand | Epidermal growth factor |
19 | AKT1 | Protein kinase | AKT serine/threonine kinase 1 |
20 | IFNG | Ligand | Interferon gamma |
21 | MAPK3 | Protein kinase | Mitogen-activated protein kinase 3 |
22 | MAPK8 | Protein kinase | Mitogen-activated protein kinase 8 |
23 | VEGFA | Ligand | Vascular endothelial growth factor A |
24 | MMP2 | | Matrix metallopeptidase 2 |
25 | BCL2 | | BCL2, apoptosis regulator |
26 | IL10 | Ligand | Interleukin 10 |
2.3 Disease similarity data
One of the problems that arises when examining diseases is the use of different names or identifiers for the same disease. In the case of the disease in the current research, bovine mastitis was used for dairy cattle, and mastitis was used for human and other mammals in the literature. The Pathway Studio web tool 12.0.1.5 was used to construct a network of disease or cell processes that were functionally associated with mastitis or bovine mastitis. Pathway Studio as a pathway analysis tool incorporates some commercial and public databases such as BIND [29], KEGG [30], and GO [30] that utilizes the ResNet Mammal database. Moreover, it also uses the powerful text-mining tool MedScan to seek the latest information from PubMed and other public sources (Elsevier-Ariadne Genomics, Rockville, MD) [31]. For more confidence, all relationships which were reported by more than two references were selected. All relations between mastitis or bovine mastitis with other diseases or cell process are indicated in Table 3. Additional details and references are provided in Additional file 1. As shown, most of the cases related to mastitis or bovine mastitis are the same and demonstrated the similarity of this disease in all mammals. This information has been added to disease similarity part of dataset shown in Table 1.
Table 3
Known disease or cell process related to mastitis or bovine mastitis by using Pathway Studio web tool (based on at list two references)
| | Related disease to mastitis or Bovine mastitis | | | Related disease to mastitis (continue) | | | Related disease to mastitis (continue) |
1 | Bovine mastitis | bacterial infection | | 34 | injury | | 67 | innate immune response |
2 | | cellular immune response | | 35 | metabolic disorder | | 68 | ketosis |
3 | | cryptococcosis | | 36 | obesity | | 69 | lactation |
4 | | Escherichia coli infection | | 37 | pain | | 70 | life span |
5 | | fever | | 38 | sepsis | | 71 | lipid degradation |
6 | | fibrosis | | 39 | skin disease | | 72 | lupus erythematosus profundus |
7 | | infection | | 40 | swelling | | 73 | maedi |
8 | | inflammation | | 41 | systemic lupus erythematosus | | 74 | milk production |
9 | | inflammatory disease | | 42 | tuberculosis | | 75 | MRSA infection |
10 | | protothecosis | | 43 | type 1 diabetes | | 76 | mumps |
11 | | staphylococcal infection | | 44 | apoptosis | | 77 | neoplasm |
12 | | milk production | | 45 | breast cancer | | 78 | neutrophil recruitment |
13 | Mastitis | abscess | | 46 | breast pain | | 79 | nipple discharge |
14 | | angiogenesis | | 47 | breast-feeding | | 80 | ovarian follicle development |
15 | | autoimmune disease | | 48 | contagious agalactia | | 81 | ovary function |
16 | | bacterial infection | | 49 | diabetes mellitus | | 82 | ovulation |
17 | | benign breast disease | | 50 | endemic disease | | 83 | parity |
18 | | breast abscess | | 51 | energy homeostasis | | 84 | parturient paresis |
19 | | cancer | | 52 | fat necrosis | | 85 | parturition |
20 | | candidiasis | | 53 | fatty liver | | 86 | pregnancy |
21 | | cattle disease | | 54 | fertilization | | 87 | proteolysis |
22 | | cell count | | 55 | fibrosis | | 88 | pseudotuberculosis |
23 | | death | | 56 | galactorrhea | | 89 | respiratory tract infection |
24 | | edema | | 57 | Gram-negative bacterial infection | | 90 | retained placenta |
25 | | endotoxemia | | 58 | granulomatous mastitis | | 91 | smoking |
26 | | Escherichia coli infection | | 59 | hypocalcemia | | 92 | staphylococcal infection |
27 | | fever | | 60 | IgG4-related disease | | 93 | subfertility |
28 | | hyperemia | | 61 | immune response | | 94 | virulence |
29 | | immunity | | 62 | immune system activation | | 95 | virus infection |
30 | | immunopathology | | 63 | immune system function | | 96 | weaning |
31 | | infection | | 64 | inbreeding | | 97 | wounds and injuries |
32 | | inflammation | | 65 | infectious disease | | | |
33 | | inflammatory disease | | 66 | inflammatory response | | | |
2.4 Drugs disease
With a review of the literature, we were able to develop a comprehensive list of drugs or antibiotics that have been used to treat E. coli mastitis (see Table 4). This information has been added to drug-disease relation part of dataset shown in Table 1.
Table 4
List of known drugs or antibiotics reported in literature to treat E. coli mastitis
| Drug or Antibiotic | Reference |
1 | Ampicillin | [32] |
2 | Aspirin | [33] |
3 | Ceftazidime | [32] |
4 | Cephalexin | [32] |
5 | Cephapirin ( Cefoperazone, Ceftiofur, Cefquinome ) | [34] |
6 | Chloramphenicol | [35] |
7 | Cinoxacin | [36] |
8 | Ciprofloxacin | [32, 36] |
9 | Dexamethasone | [37] |
10 | DHS (dihy- drostreptomycin sesquisulfate sa) | [32] |
11 | Flunixin meglumine | [38] |
12 | Fluoroquinolones (enrofloxacin, danofloxacin, marbofloxacin ) | [34] |
13 | Gentamicin | [32, 35] |
14 | Isoflupredone acetate | [33] |
15 | Ketoprofen | [32] |
16 | Meloxicam | [39] |
17 | Oxytetracycline | [40] |
18 | Penethamate hydriodide | [39] |
19 | Polymixin | [41] |
20 | Prednisolone | [42] |
21 | Tetracycline | [32] |
22 | Trimethoprim | [32] |
23 | Sulfadoxine | [40] |
24 | Sulfamethoxazole | [35] |
25 | Sulfadiazine | [32] |
2.5 Integration of data
The final heterogeneous network model was constructed by integration of these four mentioned data which were discussed in previous sections. It is necessary to mention that Heter-LP input data could be incoincident in different parts of the heterogeneous network. This means that a complete list of drugs, diseases and proteins/genes (as targets) will be achieved by union of similar typed items in each sub-network.