Datasets

[meninonas]

Dr. Yoo,

Find the merged dataset below. It has dimensions of 463x2140. I'll run the correlations on Monday and run Banjo.

[cwyoo]

Dr. Yoo,

I have coded the datasets according to Mutation, Amplification, and Expression. I was not able to merge them in one large dataset due to LibreOffice Calc not being able to open the original dataset in its entirety. LibreOffice Calc gives me the following error:

Code: Select all

The data could not be loaded completely because the maximum number of columns per sheet was exceeded.

I would suggest opening the datasets in Microsoft Office if you have it available to you.

I am currently working on labeling the genes according to _MUT, _AMP, and _EXP. Those should be ready tomorrow.

Please use the following reference coding for expression:
0: Down
1: (NULL)
2: UP

Also create the following five binary variables: Luminal_A, Luminal_B, Basal_like, HER2_enriched, Normal_like

Encode each case of the above variable 1 if PAM50_Subtype case is the same as the variable name; otherwise encode it as 0.

Select 900 gene expressions that are correlated with either PAM50_Subtype or NRF1. Analyze with banjo using 906 variables (900 gene expressions, NRF1 gene expression, Luminal_A, Luminal_B, Basal_like, HER2_enriched, Normal_like).

I have put Office 2007 ISO image file in /cdrom so mount it in your Windows virtual machine and use the key (in /cdrom as well) to install the program.

[meninonas]

Professor,

In an email, you asked me to include the following variables:

Diagnosis Age
ER Status
HER2 Final Status
Overall Survival Status
Overall Survival (Months)
PAM50 Subtype

In your last post, you don't mention ER Status, HER2 Final Status, Overall Survival Status, or Overall Survival (Months). Do you want me to delete those and separate PAM50 Subtype? Or do you want me to keep those?

I added the dataset you sent me for reference.

[meninonas]

Professor,

I have coded as following

Clinical Variables: If it equals to the corresponding Variable=1, Other=0

Genes: Down=0, (NULL)=1,UP=2

I have added the coded data. I am currently finding the correlated variables.

Once you let me know about which clinical variables too keep/delete variables I'll start running Banjo.

[cwyoo]

Professor,

I have coded as following

Diagnosis Age: >50=1, <50=0

ER Status: Negative=0, Positive=1, Otherwise=2

HER2 Final Status: Negative=0, Positive=1, Otherwise=2

Overall Survival Status: DECEASED=0, LIVING=1

Overall Survival Months: <Average=0, >Average=1

Genes: Down=0, (NULL)=1,UP=2

I have added the coded data. I am currently finding the correlated variables.

Once you let me know about which clinical variables too keep/delete variables I'll start running Banjo.

As I stated in the previous message and told you today, please use the following variables in banjo:

Select 900 gene expressions that are correlated with either PAM50_Subtype or NRF1. Analyze with banjo using 907 variables (900 gene expressions, NRF1 gene expression, Age, Luminal_A, Luminal_B, Basal_like, HER2_enriched, Normal_like).

[cwyoo]

Dr. Yoo,

Please find the reduced dataset below. I'm already working on the banjo analysis.

Please post intermediate files especially the ones with correlation calculations.

[cwyoo]

Dr. Yoo,

Please find the datasets below with the correlations. I did the following:

I took the absolute value of the correlations and ordered the dataset according to the correlations

Afterwards, I chose the top 220 variables with the highest correlations from each individual dataset and placed them within one large dataset

Finally, I deleted all of the variables that were repeated and ended with 907 variables.

I already ran banjo for one hour and have sent you the results. I already started running for two hours. I'll send you the results tomorrow morning.

Most of the csv files contain all the same entries in rows. I ask you to delete those files and upload the files with correlations information.

[meninonas]

Dr. Yoo,

Yes. The correlations are at the end of the file. I did it the same way you showed me.

[cwyoo]

Dr. Yoo,

Yes. The correlations are at the end of the file. I did it the same way you showed me.

Luis, if the first column contains all same name, how would we know what correlation you are talking about? Also, would you want to see a dataset that has all the same entries except the last column? Please think through before you post a dataset.

[cwyoo]

Professor,

Find the correlated data below. I have also sent you the 2-Hour Banjo results. I am currently running the 4-Hour Banjo Analysis.

The original data has 463 cases, but the data that you attached here has 464 cases. Why is this the case? Also post the sorted genes by correlation value per variable.