Dear all,
I am performing a meta-analysis of multiple GWAS data and now, after an intensive QC, I have to choose a suitable software for doing it. From literature it seems that the most used tool is ´Metal´ but also ´GWAMA´ is widely used. From your experience, which one performs better? Is there any other software you can suggest me? Thanks in advance for all your answers.
See figure 2 here:
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3351172/?report=classic
Appears that at the time, METAL was quite popular and likely still is. I use GWAMA, mainly due to it being popular in our lab, and have not encountered any issues with it. However, I know may other labs that also use METAL, or R packages, the later usually for analysing few variants.
It's mostly a matter of choice. METAL is the most commonly used tool. If you prefer R, rmeta is quite convenient to use. Performing the analysis using PLINK has the advantage that you can use the same program for primary and meta-analysis. I'd recommend to test at least the above three and then select the one that provides the requested output and is most convenient to use. (one should afford the time to get to know more than one)
Depends on what you want to do. If you're analyzing a half-dozen studies, just flip a coin to pick your software.
METAL and GWAMA are both good programs. METAL does not do random-effects models as far as I know. GWAMA will spit out a lot of data errors and inconsistencies, which is great for QC. I haven't used R packages for meta-analyses mainly because of limited memory and sloooow load times. If you're doing trans-ethnic meta-analyses, MANTRA is excellent--if you have access to a lot of computing power and/or patience.
I haven't used PLINK for metas, but it's worth a try since PLINK does everything well and, importantly, efficiently. PLINK2 (PLINK1.9) is a brilliant implementation of PLINK and amazingly incredibly fast. Plus, by far the best documentation in statistical genetics software, IMO. Thanks, Shaun Purcell.
My Research area: emergency; caring. qualitative and quantitative study
My Research area: emergency; caring. qualitative and quantitative study
METAL is the most commonly used tool. If you prefer R, rmeta is quite convenient to use. Performing the analysis using PLINK has the advantage that you can use the same program for primary and meta-analysis. I'd recommend to test at least the above three and then select the one that provides the requested output and is most convenient to use. (one should afford the time to get to know more than one)
METAL is good but limited, it does do Random effect. I will advise you to use Metasoft
- Badges
- Science topic
- Similar topics
- Filing