Hello everyone, I want to learn a new programming language. The purpose is to write software for animal quantitative genetics. I found that the three most popular software (BLUPF90, DMU, ASREML) in the quantitative genetics of animals are all written in Fortran. But some of my friends suggested that I learn python.
I am good at the R language. I think there is not much difference between R and Python, and Julia is also one of my languages to consider.
So I am now confused about which language to learn? Which language is more suitable for animal quantitative genetics? Anyone who knows, please give me some suggestions. I still have three years to learn this new language.
I think the easiest language is the m-file script of the Matlab. It is even higher than the python or the C language.
When you use the Matlab you can convert it into a number of lower languages using translators.
Best wishes
I think the easiest language is the m-file script of the Matlab. It is even higher than the python or the C language.
When you use the Matlab you can convert it into a number of lower languages using translators.
Best wishes
Abdelhalim abdelnaby Zekry Dear, thank you for your suggestion. I don't know Matlab very well. But my original purpose was to learn a programming language. This way I can write software. But this language must be efficient. I checked Matlab and it may not be efficient enough for large matrix algebra operations. This is just my personal opinion.
Dear Yansen Chen
Wish you success,
The meaning of matlab is the matrix laboratory. So is sought originally for solve matrices. Please see the link: https://cimss.ssec.wisc.edu/wxwise/class/aos340/spr00/whatismatlab.htm#:~:text=The%20name%20MATLAB%20stands%20for,with%20input%20from%20many%20users.
Best wishes
Dear Yansen Chen I an not from you area of expertise, but if you have time and energy then I suggest you learn Fortran (I am C/C++ programmer, but Fortan is great for matrix manipulation). I think that the main issue with Python, Matlab, R etc. is RAM and the representation of the numbers in your matrix.
Dear Abdelhalim abdelnaby Zekry
Thanks for your shared the link to Matlab. I have checked it. Maybe as Tatjana Jakšić Krüger said, the RAM is also a very important factor to me. Also, I need parallel operations to speed up matrix algebra operations.
Dear Tatjana Jakšić Krüger
Thank you for your reply. As I said in the question, Currently the main software in my field is written in Fortran. But due to the huge Python library, more and more people start to use Python. Will the software based on matrix algebra processing also migrate to Python?
Yansen Chen I am not sure if the software based on matrix algebra processing will migrate from Fortran. I think that Fortran will not be abandoned. However, the NumPy library is huge and used by many researchers so there might be two worlds existing.
Python programing is good for animal quantitative analysis
secondly R is also good tool
Jeyarani Milton
forgive me, I feel compulsed to ask why do you consider Java easy to learn. Just start wtih example of "Hello word" and compare it against Python. My most favorite is reading files with Java. I think it is a splendid example of how to complicate things that are simple in other languages.
Hi, Tatjana Jakšić Krüger ， Thank you for your share ideas. I know the Numpy and Pandas in Python are good libraries. Maybe it's time to learn Fortran. These days I check the information online, the Fortran is a better choice for me.
Hi, Ilaria Lavagi , Adnan Majeed , Thank you for your reply. Python and R both are good languages. In data preprocessing, I mainly use R and occasionally Python. But this time I want to write software that deals with linear mixed models. It mainly deals with matrix algebra and must be fast, and RME should not take up too much. So Fortran may be my best choice.
It depends from some factors:
1) how many hw recources you have
2) what is the expected performance of your system
3) how much effort you want to put in sw development
4) amount of data to be processed
5) is it a field of simulation or a data aggregation, or both
Now some hints
1) "R" software + C language, both, free, huge amount of libraries and good support from developers, very good performances on laptop, capable to interface with databases, can also use Spark libraries to have a good interface towards big data systems, mostly used for simulation and data aggregation (max 1 or 2 tera bytes of data)
2) Spark + Hadoop environemnt interfaced with some C/Pythin scripts, in this case you jump on another level mostly focused on data aggregation, poorly indicated for simulation
Adding the two solutions (1+2) then you have a playground to play as you like building what you nedd as a LEGO structure
3) you could also take alook on dated but always very good solution to be built from scratch based on Genetic Algorythms that are very weel suite for "genetic evolution " of mathematical models, so if you have already a very bsic math model you can develope a fast GA to see how the parameters of your "math model" can evolve to suite the real environment and then slowly make the model more mature and complex
Hi Zachariyah Conteh Adnan Majeed , Thank you for your answer. I also used the R to deal with the data, but it need too much RAM. So I need to to learn a new language.
Hi Alessandro Fortunato I generally analyze data on the Linux system. What I do myself is the animal quantitative genetics and breeding and the mixed equations that need to be solved. The data is generally in the hundreds of thousands. I only know R and a small amount of python myself. Thank you for your reply.
I suggest you study the article entitled ''Compare six animal breeding software packages'' by Rostam Abdollahi et al., 2013, which compares different types of software and their advantages and disadvantages.
Paper link: https://ijas.ut.ac.ir/article_29346.html?lang=en
Generally, It depends on the purpose of your research.
For example, DMU software is common and appropriate for variance components and breeding value estimation.
Mohammad Ghaderzadeh I checked the article you recommended, but sorry, I can’t understand the Iranian language. I am also writing software that can calculate the variance components of different special models and get genetic correlations.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Programming Languages