SMLG (Statistical Machine Learning Group) Discussion Forum

Good evening everyone,
Here is my dataset I've been building based on my research of tumor-like proliferative vascular diseases which is still not finalized for analysis. Thank you.

Dr. Yoo,
Here are the demographics of the multiple data-sets with one of the diseases IPAH. It includes the description of the data-set, number of subjects and gender, ages, ethnicity if available, and the platform used. I will post all the other disease demographics plus the data-sets in their original form once completed. Thanks.

Dr. Yoo & Dr. Felty,
Attached are the rest of the demographics of the remaining datasets. Abbreviations are as followed: ATHER (atherosclerosis), LAM (Lymphangioleiomyomatosis), PDR (Proliferative Diabetic Retinopathy), IPAH (idiopathic pulmonary arterial hypertension), and IH (Infantile hemangioma). It should be demographics for a total of 20 datasets all together. I will post all the data once it is all curtailed the way Dr. Yoo requested. Thanks.

Dr. Yoo,
Attached are one of the datasets from one of the atherosclerosis studies. I wanted to make sure that I organized the data in the correct order before I start posting the rest of them. Can you please let me know if I have to make any changes to the formatting. Thanks.

vavec001 wrote:Dr. Yoo,
Attached are one of the datasets from one of the atherosclerosis studies. I wanted to make sure that I organized the data in the correct order before I start posting the rest of them. Can you please let me know if I have to make any changes to the formatting. Thanks.

Looks ok. among diffrent files, you need to identify common genes among them. please report the number of common genes and number of subject as well.

Dr. Yoo,
Attached are the datasets for the diseases as followed: Proliferative Diabetic Retinopathy (PDR), Infantile Hemangioma (IH), and Arteriovenous Malformation (AVM) in .csv format. I will also post the datasets for arteriosclerosis and IPAH as well. Thank you.

Afternoon Dr. Yoo,
Below is a listing of the common genes, number of samples and cases/controls. I am still having trouble merging all the different GPL files together but concluded the number of common genes based on their platforms. The attached file is tab delimited for each GPL which contains the expressed genes.

GPL570: 8 datasets, 21025 common genes, 147 samples, and 77 cases/70 controls
GPL 1708: 2 datasets, 18841 common genes, 35 samples, and 19 cases/ 16 controls
GPL 96: 2 datasets, 13211 common genes, 28 samples, and 16 cases/ 12 controls
GPL 6884 2 datasets,18196 common genes, 13 samples, and 8 cases/5 controls
GPL OTHERs: 9 datasets, 350 samples, common genes & cases/controls: TBD
Total Datasets: 22 datasets

Dr. Yoo,
I was able to identify the common genes amongst all the datasets which the total number was 4280 genes. I attached an excel file with the listing of expressed genes plus total number of cases vs. controls and number of samples. Thank you.

Dr. Yoo,
I had previously grouped the same GPLs together to find the common genes amongst all the datasets. In total when I recounted it was 6 diseases, 24 datasets, 304 cases / 189 controls, and 689 samples. Attached is the file with cases and controls for each dataset. I will post a excel file with the demographics and variables for each of the datasets. Thanks.

Attached are two heat-maps generated from R concerning a dataset from this manuscript
Identification of epigenetic regulators in the array data from the paper, Stem Cell Reports. 2015 Oct 20 Induced Developmental Arrest of Early Hematopoietic Progenitors Leads to the Generation of Leukocyte Stem Cells.
http://www.sciencedirect.com/science/ar ... 111500274X

The idea was to see if there is any potential interaction with the Polycomb Complex (PRC2) with IdHP ID3 over-expressing hematopoetic cells vs. control genes.
1) I used the genes from that dataset (IdHP ID3 over-expressing hematopoetic cells vs. control) that were involved in the Polycomb Complex (PRC2) such as EZH1, EZH2, EED, SUZ12, and the key gene ID3
2) I came across a review that was published in Nature (2011) about how although the core of the complex is EZH1/2, SUZ12, EED, and RBAP46/48 it does however have more genes with various functions such as AEBP2, PCLs, and JARID2. Furthermore it also has ones that interact with the PRC2 (DNMTs, HDAC1, and SIRT1) but their effect on the PRC2 is unclear so I furthermore used this in the second heat-map generated.

Thank you