Class Journal Week 8 - Revision history

Nanguiano: /* Nicole Anguiano */ Fix a formatting mistake

2016-09-16T23:32:41Z

‎Nicole Anguiano: Fix a formatting mistake

Nanguiano: /* Nicole Anguiano */ Added answers

2016-09-16T23:32:11Z

‎Nicole Anguiano: Added answers

Nanguiano: Add questions and title

2016-09-16T21:05:20Z

Add questions and title

Lenaolufson: added my shared journal assignment

2015-10-27T17:16:17Z

added my shared journal assignment

Nanguiano: oops

2015-10-27T16:45:29Z

oops

@@ Line 174: / Line 174: @@
 #What recommendations does Dr. Baggerly recommend for reproducible research? How do these correspond to what DataONE recommends?
 #* Dr. Baggerly recommends that scientists provide data and code, have a system of labeling that is clear and consistent, keep a thorough description of steps that are not performed in code, and to have descriptions of planned designs. These correspond well to what DataONE recommends.
-#* Do you have any further reaction to this case after viewing Dr. Baggerly's talk?
+# Do you have any further reaction to this case after viewing Dr. Baggerly's talk?
-#** After watching the talk, I'm still dismayed that this case managed to get so far before trials were shut down. This talk highlighted how many times Baggerly and Coombs tried to get answers for why the data was incorrect, and how many times they were blown off or ignored. It also is slightly worrying to see that the recommendations for reproducible research are only recommendations and not industry standard following a mistake of that scale. It worries me that is possible for something like this to happen again, but I am glad to see that there are those who will fight for reproducible research to become the standard.
+#* After watching the talk, I'm still dismayed that this case managed to get so far before trials were shut down. This talk highlighted how many times Baggerly and Coombs tried to get answers for why the data was incorrect, and how many times they were blown off or ignored. It also is slightly worrying to see that the recommendations for reproducible research are only recommendations and not industry standard following a mistake of that scale. It worries me that is possible for something like this to happen again, but I am glad to see that there are those who will fight for reproducible research to become the standard.
 #Go back to the Merrell et al. (2002) paper and look at your "sanity check" where you compared the fold changes and p values for certain genes between your work and the paper. Did the values match? Why do you think that is? Do you think there is sufficient information there for you to reproduce their data analysis? Why or why not?
 #*The values do not match exactly, likely because the software used was different. If we both had used the exact same software, it is likely that we would have obtained the same results. With the information given in the paper, I'm not fully sure I'd be able to reproduce their data analysis even if I had the software. They don't fully explain what they did or why they did it, which would make it difficult to perform their experiment again.

@@ Line 171: / Line 171: @@
 ==Nicole Anguiano==
 #What were the main issues with the data and analysis identified by Baggerly and Coombs? What best practices enumerated by DataONE were violated? Which of these did Dr. Baggerly claim were common issues?
-#*
+#* The main issues with the data and analysis were mislabeling, gene duplication, and off by one errors. The data violated the best practices of maintain dataset provenance, review metadata of context, methods, and meaning, and use reproducible workflows. Dr. Baggerly claimed that the simplest issues were most common, particularly the issues of mixing up labels and incomplete documentation.
 #What recommendations does Dr. Baggerly recommend for reproducible research? How do these correspond to what DataONE recommends?
-#*
+#* Dr. Baggerly recommends that scientists provide data and code, have a system of labeling that is clear and consistent, keep a thorough description of steps that are not performed in code, and to have descriptions of planned designs. These correspond well to what DataONE recommends.
 #Go back to the Merrell et al. (2002) paper and look at your "sanity check" where you compared the fold changes and p values for certain genes between your work and the paper. Did the values match? Why do you think that is? Do you think there is sufficient information there for you to reproduce their data analysis? Why or why not?
-#*
+#*The values do not match exactly, likely because the software used was different. If we both had used the exact same software, it is likely that we would have obtained the same results. With the information given in the paper, I'm not fully sure I'd be able to reproduce their data analysis even if I had the software. They don't fully explain what they did or why they did it, which would make it difficult to perform their experiment again.
 [[Category:Journal Entry]]
 [[Category:Shared]]

@@ Line 168: / Line 168: @@
 [[User:Lenaolufson|Lenaolufson]] ([[User talk:Lenaolufson|talk]]) 10:16, 27 October 2015 (PDT)
 [[Category:Journal Entry]]
 [[Category:Shared]]

@@ Line 156: / Line 156: @@
 #*The values did not entirely match up (many genes that were shown to have significant changes in expression, in the paper, were found to not have any significant changes in the work related to the assignment). The differences between the paper and the work done for this week could be the result of differences in the methods related to the statistical analysis. Merrell et al. utilized a two-class SAM analysis while we utilized Excel and the calculation of the p-values of the log fold changes that reflected all of the samples. Looking at just the information in the paper, it is not exactly clear as to how somebody would reproduce their results; the methods and the rationale behind the statistics is not very clearly touched upon. However, even if the article is fairly light on some of the statistics, I think that through some additional independent research (and data hunting, since the article contains dead links) one would be able to get a result that is similar to what was found in the paper. It seems that this article is chiefly written with professionals in the scientific community (specifically, those involved with genes/microarrays) as the intended audience, who are likely better versed in the statistics and the programs related to this kind of research.
 [[User:Blitvak|Blitvak]] ([[User talk:Blitvak|talk]]) 02:06, 27 October 2015 (PDT)
 [[Category:Journal Entry]]
 [[Category:Shared]]

@@ Line 10: / Line 10: @@
 [[User:Msaeedi23|Msaeedi23]] ([[User talk:Msaeedi23|talk]]) 23:38, 26 October 2015 (PDT)
-== Nicole Anguiano ==
+<!--== Nicole Anguiano ==
 #What were the main issues with the data and analysis identified by Baggerly and Coombs? What best practices enumerated by DataONE were violated? Which of these did Dr. Baggerly claim were common issues?
 #*
@@ Line 20: / Line 20: @@
 #*
-[[User:Nanguiano|Nanguiano]] ([[User talk:Nanguiano|talk]]) 15:06, 20 October 2015 (PDT)
+[[User:Nanguiano|Nanguiano]] ([[User talk:Nanguiano|talk]]) 15:06, 20 October 2015 (PDT)-->