Tumor-Like Proliferative Vascular Disease Dataset

Currently undergoing other projects. Once there is significant traction, we will create new forums for those analyses and results.

Re: Tumor-Like Proliferative Vascular Disease Dataset

Postby vavec001 » Mon Dec 12, 2016 11:31 am

Dr Yoo,
I ran banjo at multiple time points (1, 3, and 6 hour) successfully but when I tried to direct banjo to produce the .dot image it kept on giving me a response "segmentation fault (core dumped)". Will continue to figure out how to fix this problem but listed below are the network scores from each run. Thanks.

1 hour:
Network score: -1139509.3927, first found at iteration 80977
3665 (at 20 mins)
Network score: -1079957.5136, first found at iteration 116993
3665 (at 40 mins)
Network score: -1053337.8670, first found at iteration 150985
3665 (at 1 hour)
Network score: -1053337.8670, first found at iteration 150985
3665 (Best Network Overall)

3 hour:
Network score: -1145208.4305, first found at iteration 79968
3665 (at 20 mins)
Network score: -1082527.6755, first found at iteration 115987
3665 (at 40 mins)
Network score: -1052393.3835, first found at iteration 150952
3665 (at 1 hr)
Network score: -1037215.8297, first found at iteration 181990
3665 (at 1.34 hr)
Network score: -1037215.8297, first found at iteration 181990
3665 (at 1.67 hr)
Network score: -1037215.8297, first found at iteration 181990
3665 (at 2 hr)
Network score: -1037215.8297, first found at iteration 181990
3665 (at 2.34 hr)
Network score: -1029636.5662, first found at iteration 336983
3665 ( at 2.67 h)
Network score: -1029636.5662, first found at iteration 336983
3665 (at 3 hr)
Network score: -1029636.5662, first found at iteration 336983
3665 (Best network overall)

6 hour:
Network score: -1147504.9399, first found at iteration 80000
3665 (at 20 mins)
Network score: -1081219.1039, first found at iteration 115940
3665 (at 40.32 mins)
Network score: -1049428.1221, first found at iteration 151994
3665 (at 1.01 hr)
Network score: -1034951.2184, first found at iteration 182996
3665 (at 1.33 hr)
Network score: -1034951.2184, first found at iteration 182996
3665 (at 1.67 hr)
Network score: -1034951.2184, first found at iteration 182996
3665 (at 2 hr)
Network score: -1034951.2184, first found at iteration 182996
3665 (at 2.33 hr)
Network score: -1028914.5366, first found at iteration 337982
3665 (at 2.67 hr)
Network score: -1028914.5366, first found at iteration 337982
3665 (at 3 hr)
Network score: -1028914.5366, first found at iteration 337982
3665 (3.33 hr)
Network score: -1028914.5366, first found at iteration 337982
3665 (3.67 hr)
Network score: -1027005.9166, first found at iteration 490970
3665 (at 4.01 hr)
Network score: -1027005.9166, first found at iteration 490970
3665 (at 4.34 hr)
Network score: -1027005.9166, first found at iteration 490970
3665 (at 4.68 hr)
Network score: -1026556.6477, first found at iteration 644963
3665 (at 5.01 hr)
Network score: -1026257.8594, first found at iteration 645973
3665 (at 5.34 hr)
Network score: -1026257.8594, first found at iteration 645973
3665 (at 5.67 hr)
Network score: -1026257.8594, first found at iteration 645973
3665 (at 6 hr)
Network score: -1026257.8594, first found at iteration 645973
3665 (Best Network Overall)
vavec001
 
Posts: 34
Joined: Thu May 28, 2015 12:49 pm

Re: Tumor-Like Proliferative Vascular Disease Dataset

Postby vavec001 » Mon Jan 30, 2017 8:46 pm

1) Worked at path 4 to try and solve the graph (.dot) problem. Ran 3665 variables - 681 samples for 5 mins to test the new setting file and it ran successfully & produced a graph in an (.svg) format which can be opened through the internet and a (.png) format. Ran banjo for 1 hour and it ran successfully but was not able to produce a graph (.dot). Attached is the network in a (.png) format that was ran for 5 mins.
graph.dot.png
graph.dot.png (1.45 MiB) Viewed 4939 times


2) Have been trying to use Jigan's code in R but unsuccessful thus far. Will need to further work on it and will post datasets once more progression has been made.
Last edited by vavec001 on Mon Feb 06, 2017 9:31 am, edited 1 time in total.
vavec001
 
Posts: 34
Joined: Thu May 28, 2015 12:49 pm

Re: Tumor-Like Proliferative Vascular Disease Dataset

Postby vavec001 » Sat Feb 04, 2017 6:47 pm

1) From our commonly expressed 3665 gene dataset I identified 500 up-regulated and 500 down-regulated genes using a fold change and ran banjo for 1 hour. Network Image generated is attached.
1000updownvaso.png
500 Up- & 500 -down
1000updownvaso.png (35.63 MiB) Viewed 4940 times


2) Also identified from our 3665 gene dataset 250 up-regulated and 250 down-regulated genes using a fold change and ran banjo for 1 hour. Network image generated is attached.
top.graph.2017.02.04.14.42.26.png
250 -Up & 250 -down
top.graph.2017.02.04.14.42.26.png (5.54 MiB) Viewed 4940 times


3) Generated a ID3 model using banjo with up- and -down regulated genes that were either expressed in RNA-sequenced data of cells over-expressing ID3 versus vector alone or ID3 over-expressing cells treated with PCB153 24hr vs control from lab. Identified what genes were commonly expressed from the list of RNA-sequenced data across all datasets and ran banjo to see what gene-gene interactions exist in vascular diseases. Network Image generated is attached.
ID3rna.dot.jpeg
ID3_RNA
ID3rna.dot.jpeg (38.21 KiB) Viewed 4940 times


4) McNamara et al (2010) identified a single nucleotide polymorphism (SNP) in the ID3 gene that is associated with carotid intima-media thickness; a surrogate marker of atherosclerotic plaque burden and clinical cardiovascular disease events in humans. Loss of ID3 resulted in modulation of many genes in pathways in atherogenesis. Genes that were up- and -down regulated in ID3-/- smooth muscle cells were identified based on (>4-fold) change. Identified the same up- and -down regulated genes from the ID3 -/- model in the study that were also expressed in our vascular disease dataset and ran banjo.
top.graph.2017.02.04.15.50.09.png
ID3_SNP
top.graph.2017.02.04.15.50.09.png (100.97 KiB) Viewed 4940 times


5) In the midst of running the same model from 3 and 4 in Bene. Will post results once they are available.
vavec001
 
Posts: 34
Joined: Thu May 28, 2015 12:49 pm

Re: Tumor-Like Proliferative Vascular Disease Dataset

Postby vavec001 » Mon Feb 20, 2017 10:01 am

We identified the expression levels of ID genes (ID1, ID3, & ID4) that were expressed across all GEO data sets alongside commonly expressed genes from RNA-sequence data of cells over-expressing ID3 versus vector alone & RNA-sequence data of ID3 over-expressing cells treated with PCB153 24h versus control. 13 genes were identified (ECM2, EGR1, CD4, GPX2, COL3A1, COL1A2, COL4A3, BGN, GRIN2D, HES1, ESRRG, KCNJ2, IGFBP5). Furthermore looked at ID3 regulated inflammatory factors expressed in the data and we identified 3 genes thus far (IL5, IL6, & IL10). Data is divided into tissue samples versus blood samples.
Attachments
IDGenes_GEO_2-20-17.xlsx
(18.77 KiB) Downloaded 167 times
vavec001
 
Posts: 34
Joined: Thu May 28, 2015 12:49 pm

Re: Tumor-Like Proliferative Vascular Disease Dataset

Postby vavec001 » Fri Mar 03, 2017 7:49 pm

This week's update:
1)ID3 correlation tests with common gene dataset (divided into blood & tissue samples)
ID3correlation_commongenes_tissue&bloodsamples.xlsx
(213.5 MiB) Downloaded 173 times

2) ID3 correlation tests with common genes within each vascular disease (tissue samples: LAM: Lymphangioleiomyomatosis, PDR: proliferative diabetic retinopathy, AVM: arteriovenous malformation, IH: Infantile Hemangioma, IPAH: Idiopathic pulmonary arterial hypertension)
ID3-correlation_tissuesamples.xlsx
(327.78 KiB) Downloaded 163 times

3)ID3 correlation with expressed ID family partners (ID1,ID4). Divided by all blood and tissue samples & for each vascular disease (blood and tissue samples).
ID protein correlation tables_3-3-17.docx
(96.99 KiB) Downloaded 190 times


Still need to complete:
1) ID3 correlation tests with common genes within each vascular disease (blood samples)
2) Need to compare gene lists from Dr. Felty and from literature review to see which are common
3) Start to generate and complete heat-map and look over and complete demographic tables
vavec001
 
Posts: 34
Joined: Thu May 28, 2015 12:49 pm

Re: Tumor-Like Proliferative Vascular Disease Dataset

Postby vavec001 » Mon Mar 27, 2017 10:54 am

1) Table of GEO datasets included
2) 3665 common genes identified across all datasets
3) a. Did a ID3 correlation of all common genes in blood samples and calculated the Quartiles (Used quartiles 3 & 4)
b. Did a ID3 correlation of all common genes in tissue samples and calculated the Quartiles (Used quartiles 3 & 4)
c. Did a ID3 correlation of all common genes in specific vascular diseases (1. PAD -peripheral arterial disease, 2. LAM - Lymphangioleiomyomatosis, 3. IH - Infantile Hemangioma, 4. ATHER- atherosclerosis, 5. IPAH - idiopathic pulmonary arterial hypertension, 6. AVM - Arteriovenous Malformation, 7. PDR - Proliferative Diabetic Retinopathy)
4) Performed an independent t-test (Gene expression in blood samples, tissue samples, specific vascular diseases) and calculated the Quartiles (Used Q1)
5) Genes that fell in the ID3 correlation of Quartile 3 & 4 and Quartile 1 of the independent t-test were matched. Results attached with correlation and p-value
Attachments
ID3correlation_finalALL_3-27-17.xlsx
(157.85 KiB) Downloaded 163 times
GEO_Table_3-27-17.xlsx
(10.83 KiB) Downloaded 196 times
vavec001
 
Posts: 34
Joined: Thu May 28, 2015 12:49 pm

Re: Tumor-Like Proliferative Vascular Disease Dataset

Postby vavec001 » Fri Apr 21, 2017 4:50 pm

Bene Structure created with common ID3-correlated blood & tissue genes with a location variable and vascular variable
Attachments
iris.png
iris.png (441.63 KiB) Viewed 4940 times
vavec001
 
Posts: 34
Joined: Thu May 28, 2015 12:49 pm

Re: Tumor-Like Proliferative Vascular Disease Dataset

Postby vavec001 » Sun Apr 23, 2017 10:47 am

Model of best fit with ID3-common genes in blood & tissue in Genie
Attachments
Untitled.png
Untitled.png (40.88 KiB) Viewed 4940 times
vavec001
 
Posts: 34
Joined: Thu May 28, 2015 12:49 pm

Re: Tumor-Like Proliferative Vascular Disease Dataset

Postby vavec001 » Mon May 01, 2017 10:13 am

ID3-common genes in blood & tissue in various states
1) blood
Blood.PNG
Blood.PNG (32.55 KiB) Viewed 4940 times

2) tissue
tissue.PNG
tissue.PNG (33.31 KiB) Viewed 4940 times

3) disease
disease.PNG
disease.PNG (32.87 KiB) Viewed 4940 times

4) control
control.PNG
control.PNG (33.21 KiB) Viewed 4940 times

5) ID3 normal expression
ID3_normal.PNG
ID3_normal.PNG (33.21 KiB) Viewed 4940 times

6) ID3 high expression
ID3_high.PNG
ID3_high.PNG (33.31 KiB) Viewed 4940 times

7) ID3 low expression
ID3_low.PNG
ID3_low.PNG (33.85 KiB) Viewed 4940 times
vavec001
 
Posts: 34
Joined: Thu May 28, 2015 12:49 pm

Re: Tumor-Like Proliferative Vascular Disease Dataset

Postby vavec001 » Mon Oct 16, 2017 8:35 am

Using Efrain's code "RAutoCIDs.R" I was able to clean 17 of my 20 vascular disease datasets. After meeting & speaking with Efrain last week about the problems I ran into running the code on the remainder of my datasets, I will finish up the rest of the datasets and then use the R code "RMatchGenes.R" to find the common genes. Thank you.
vavec001
 
Posts: 34
Joined: Thu May 28, 2015 12:49 pm

PreviousNext

Return to Others

Who is online

Users browsing this forum: No registered users and 1 guest

cron