This preprint is under consideration at BMC Medical Genomics. A preprint is a preliminary version of a manuscript that has not completed peer review at a journal. Research Square does not conduct peer review prior to posting preprints. The posting of a preprint on this server should not be interpreted as an endorsement of its validity or suitability for dissemination as established information or for guiding clinical practice.
Learn more about In Review
Research article
Sadegh Azimzadeh Jamalkandi
Chemical Injuries Research Center, Systems Biology and Poisonings Institute, Tehran, Iran
Corresponding Author
ORCiD: https://orcid.org/0000-0003-3403-3700
License:
This work is licensed under a CC BY 4.0 License. Read Full License
Background
Circulating tumor cells (CTCs) are the critical initiators of distant metastasis formation. In which, the reciprocal interplay among different metastatic pathways which promote survival of CTCs, is not well introduced, using network approaches. CTC cells include single and cluster cells, in which cluster cells revealed 23-50 fold more metastatic potentials.
Here, to investigate the unknown pathways of single/cluster CTCs, the co-expression network reconstructed, using WGCNA (Weighted Correlation Network Analysis) method. Having used the hierarchical clustering, we detected the Immune-response and EMT subnetworks. The metastatic potential of genes was assessed and validated through the support vector machine (SVM), neural network, and decision tree methods on two external datasets. To identify the active signaling pathways in CTCs, we reconstructed a casual network. The Log-Rank test and Kaplan-Meier curve were applied to detect prognostic gene signatures for metastasis-free survival. Finally, a predictive model was developed for metastasis risk of patients, using VIF-stepwise feature selection.
Results
Our results showed the crosstalk among EMT, the immune system, menstrual cycles, and the stemness pathway in CTCs. In which, fluctuation of menstrual cycles is a new detected pathway in breast cancer CTCs. The reciprocal association between immune responses and EMT was identified in single/cluster CTCs. The SVM model indicated a high metastatic potential of EMT subnetwork (accuracy, sensitivity, and specificity scores were 87%). The distant-metastasis-free-survival model was identified to predict patients’ metastasis risks. (c-index=0.8). Finally, novel metastatic biomarkers including PTCRA, F13A1, ICAM2, and SNRPC were detected in breast cancer.
Conclusions
In conclusion, the reciprocal interplay among critical pathways in CTCs enhances their survival and metastatic potentials. Such findings may help to develop more precise predictive metastatic-risk models or detect novel biomarkers.
Keywords
breast cancer, single / cluster CTC, metastasis, co-expression, epithelial-mesenchymal transition, immune response
Figure 1
Figure 2
Figure 3
Figure 4
This preprint is available for download as a PDF.
This is a list of supplementary files associated with this preprint. Click to download.
- Supplementary.docx
- TableS1.xlsx
Table S1. midnightblue (immune) subnetwork. The list of genes, module membership, the logarithm of fold change, differential analysis adj-p.value, and gene-significance statistics.
- TableS2.xlsx
Table S2. turuise (EMT) subnetwork. The list of genes, module membership, the logarithm of fold change, differential analysis adj-p.value, and gene-significance statistics.
- TableS3.xlsx
Table S3. The preservation statistics. This table includes Z_(summary ) and 〖Median〗_rank statistics.
- TableS4.xlsx
Table S4. Biological pathways in the directed network. We categorized genes of the detected directed network, using ClueGO plugin in Cytoscape. The Biological pathways and p-values were reported in Table S4.
- TableS5.xlsx
Table S5. the EMT subnetwork cox-PH results. The cox-PH analysis was implemented for the EMT subnetwork genes. The selected genes, coefficients, and p-values were reported in Table S5.
- TableS6.xlsx
Table S6. the Immune subnetwork cox-PH results The cox-PH analysis was implemented for the immune subnetwork genes. The selected genes, coefficients, and p-values were reported in Table S6.
- TableS7.xlsx
Table S7. EMT VIF values. To investigate multicollinearity in the cox-PH model, we calculated the Variance Inflation Factor (VIF). The VIF < 10 indicates no multicollinearity. The VIF of immune genes was reported in Table S7.
- TableS8.xlsx
Table S8. immune VIF values. To investigate multicollinearity in the cox-PH model, we calculated the Variance Inflation Factor (VIF). The VIF < 10 indicates no multicollinearity. The VIF of immune genes was reported in Table S8.
- FigureS1.tif
Figure s1. subnetwork-trait correlation p-values. The metastasis-related subnetwork was selected, using subnetwork/trait correlation. (|correlation |> 0.5)
- FigureS2.tif
Figure S2. The Schoenfeld residuals for EMT genes. The proportional hazard ratio investigated, using the Schoenfeld residuals. The residuals (red dots) must be between the curves.
- FigureS3.tiff
Figure S3. The Schoenfeld residuals for immune genes. The proportional hazard ratio investigated, using the Schoenfeld residuals. The residuals (red dots) must be between the curves.
Loading...
Badges
Prescreen
This manuscript passed our Prescreen assessment
Complete author information
Appropriate declaration statements
Potential risks to human health
Prescreen
Peer Review Timeline
CURRENT STATUS: Under Review
Version 1
Posted 21 Sep, 2020
No community comments so far
Editorial decision: Major revision
On 23 Nov, 2020
Review #3 received
Received 07 Nov, 2020
Review #2 received
Received 07 Nov, 2020
Reviewer #3 agreed
On 29 Oct, 2020
Reviewer #2 agreed
On 28 Oct, 2020
Review #1 received
Received 28 Oct, 2020
Reviewer #1 agreed
On 07 Oct, 2020
Reviewers invited
Invitations sent on 03 Oct, 2020
Editor assigned
On 24 Sep, 2020
Submission checks complete
On 18 Sep, 2020
Editor invited
On 18 Sep, 2020
First submitted
On 10 Sep, 2020
Metrics
Comments: 0
PDF Downloads: ...
HTML Views: ...
Subject Areas
Epigenetics & Genomics
Learn more about our company.
This preprint is under consideration at BMC Medical Genomics. A preprint is a preliminary version of a manuscript that has not completed peer review at a journal. Research Square does not conduct peer review prior to posting preprints. The posting of a preprint on this server should not be interpreted as an endorsement of its validity or suitability for dissemination as established information or for guiding clinical practice.
Learn more about In Review
Research article
Sadegh Azimzadeh Jamalkandi
Chemical Injuries Research Center, Systems Biology and Poisonings Institute, Tehran, Iran
Corresponding Author
ORCiD: https://orcid.org/0000-0003-3403-3700
Badges
Prescreen
This manuscript passed our Prescreen assessment
Complete author information
Appropriate declaration statements
Potential risks to human health
Prescreen
Peer Review Timeline
CURRENT STATUS: Under Review
Version 1
Posted 21 Sep, 2020
No community comments so far
Editorial decision: Major revision
On 23 Nov, 2020
Review #3 received
Received 07 Nov, 2020
Review #2 received
Received 07 Nov, 2020
Reviewer #3 agreed
On 29 Oct, 2020
Reviewer #2 agreed
On 28 Oct, 2020
Review #1 received
Received 28 Oct, 2020
Reviewer #1 agreed
On 07 Oct, 2020
Reviewers invited
Invitations sent on 03 Oct, 2020
Editor assigned
On 24 Sep, 2020
Submission checks complete
On 18 Sep, 2020
Editor invited
On 18 Sep, 2020
First submitted
On 10 Sep, 2020
Metrics
Comments: 0
PDF Downloads: ...
HTML Views: ...
Subject Areas
Epigenetics & Genomics
License:
This work is licensed under a CC BY 4.0 License. Read Full License
Background
Circulating tumor cells (CTCs) are the critical initiators of distant metastasis formation. In which, the reciprocal interplay among different metastatic pathways which promote survival of CTCs, is not well introduced, using network approaches. CTC cells include single and cluster cells, in which cluster cells revealed 23-50 fold more metastatic potentials.
Here, to investigate the unknown pathways of single/cluster CTCs, the co-expression network reconstructed, using WGCNA (Weighted Correlation Network Analysis) method. Having used the hierarchical clustering, we detected the Immune-response and EMT subnetworks. The metastatic potential of genes was assessed and validated through the support vector machine (SVM), neural network, and decision tree methods on two external datasets. To identify the active signaling pathways in CTCs, we reconstructed a casual network. The Log-Rank test and Kaplan-Meier curve were applied to detect prognostic gene signatures for metastasis-free survival. Finally, a predictive model was developed for metastasis risk of patients, using VIF-stepwise feature selection.
Results
Our results showed the crosstalk among EMT, the immune system, menstrual cycles, and the stemness pathway in CTCs. In which, fluctuation of menstrual cycles is a new detected pathway in breast cancer CTCs. The reciprocal association between immune responses and EMT was identified in single/cluster CTCs. The SVM model indicated a high metastatic potential of EMT subnetwork (accuracy, sensitivity, and specificity scores were 87%). The distant-metastasis-free-survival model was identified to predict patients’ metastasis risks. (c-index=0.8). Finally, novel metastatic biomarkers including PTCRA, F13A1, ICAM2, and SNRPC were detected in breast cancer.
Conclusions
In conclusion, the reciprocal interplay among critical pathways in CTCs enhances their survival and metastatic potentials. Such findings may help to develop more precise predictive metastatic-risk models or detect novel biomarkers.
Figure 1
Figure 2
Figure 3
Figure 4
This preprint is available for download as a PDF.
This is a list of supplementary files associated with this preprint. Click to download.
- Supplementary.docx
- TableS1.xlsx
Table S1. midnightblue (immune) subnetwork. The list of genes, module membership, the logarithm of fold change, differential analysis adj-p.value, and gene-significance statistics.
- TableS2.xlsx
Table S2. turuise (EMT) subnetwork. The list of genes, module membership, the logarithm of fold change, differential analysis adj-p.value, and gene-significance statistics.
- TableS3.xlsx
Table S3. The preservation statistics. This table includes Z_(summary ) and 〖Median〗_rank statistics.
- TableS4.xlsx
Table S4. Biological pathways in the directed network. We categorized genes of the detected directed network, using ClueGO plugin in Cytoscape. The Biological pathways and p-values were reported in Table S4.
- TableS5.xlsx
Table S5. the EMT subnetwork cox-PH results. The cox-PH analysis was implemented for the EMT subnetwork genes. The selected genes, coefficients, and p-values were reported in Table S5.
- TableS6.xlsx
Table S6. the Immune subnetwork cox-PH results The cox-PH analysis was implemented for the immune subnetwork genes. The selected genes, coefficients, and p-values were reported in Table S6.
- TableS7.xlsx
Table S7. EMT VIF values. To investigate multicollinearity in the cox-PH model, we calculated the Variance Inflation Factor (VIF). The VIF < 10 indicates no multicollinearity. The VIF of immune genes was reported in Table S7.
- TableS8.xlsx
Table S8. immune VIF values. To investigate multicollinearity in the cox-PH model, we calculated the Variance Inflation Factor (VIF). The VIF < 10 indicates no multicollinearity. The VIF of immune genes was reported in Table S8.
- FigureS1.tif
Figure s1. subnetwork-trait correlation p-values. The metastasis-related subnetwork was selected, using subnetwork/trait correlation. (|correlation |> 0.5)
- FigureS2.tif
Figure S2. The Schoenfeld residuals for EMT genes. The proportional hazard ratio investigated, using the Schoenfeld residuals. The residuals (red dots) must be between the curves.
- FigureS3.tiff
Figure S3. The Schoenfeld residuals for immune genes. The proportional hazard ratio investigated, using the Schoenfeld residuals. The residuals (red dots) must be between the curves.
Loading...