Data collections over individual studies

Collections of medical data from many studies are better for understanding disease than single studies. Conclusions linking genetic mutations to diseases in individual studies have often been wrong due to small sample sizes. By publicly aggregating the data from many papers together the data becomes larger and more powerful. Merging and comparing cross study data takes powerful normalization tools. Luckily our data normalization tools are getting more and more powerful. One tool is ComBat to remove batch effects in studies: https://www.bu.edu/jlab/wp-assets/ComBat/Abstract.html available from Biocondcutor, https://www.bioconductor.org/ for R, https://cran.r-project.org/.

Comments

Post a Comment

Popular posts from this blog

In a great place - make a great place

Image
This is my recommendation for Swampscott's town development meeting. Swampscott is a small town in beautiful place by the ocean and I would love to be able to walk around town more. Walkable communities are healthier having lower rates of obesity and diabetes, property values are higher because people want to live in walkable communities and there aren't enough of them. To get to a walkable community we need to develop infrastructure for walking and bicycling like more sidewalks and trails. We need to build mixed use higher density living/working/shopping/recreation zones so that people live and work close to things they want to walk to. Higher density mixed use development is made by permitting taller buildings with street level shopping, office and condo living at higher levels. The higher density needs to be offset by more public green spaces which will benefit the entire community more than individual back yards. And higher density living frees up more space for public u...
Read more

PhD Generalist

A PhD degree actually makes you a generalist. This is because for the dissertation you are in charge of the total project. It turns out any project is made of multiple components. As the project leader you have to plan, order supplies, schedule, deal with people, plan experiments, execute the experiments, analyze the data, write papers, promote your work. Because as a graduate student you won't have a staff working for you, you get to do a little bit of everything. When I needed access to clinical data I had to work through the administration to get myself permission. When I needed computer hardware I had to set it up. When I needed software I either downloaded and installed it or wrote it myself. I had to collect data, measure the data, analyze the data, write papers, and then promote the work at conferences. I found at earlier levels of education I could focus exclusively on laboratory experiments or on developing computer software tools but then at the highest level of education...
Read more

Most efficient ways to build fitness

Some of the most efficient ways to exercise are weight lifting about 24 repetitions of each exercise to complete exhaustion and interval training like some kind of sprinting. Lower weight higher repetition weight lifting (meaning about 24 reps) done to complete muscle exhaustion (lift until you can't lift a single time more) can actually build muscle even faster low repetition with a heavier weight. A study on this is here: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0012033 N. A. Burd, et al. “Low-Load High Volume Resistance Exercise Stimulates Muscle Protein Synthesis More Than High-Load Low Volume Resistance Exercise in Young Men,” PLoS ONE, vol. 5, no. 8, p. e12033, 2010. Interval training works so well partly because you move fast turning on more fibers. Each muscle fiber in a skeletal muscle either contracts fully or not at all. You lift heavier weights or move faster by contracting/engaging more muscle fibers. So fast motions in interval training...
Read more