I know how to do NGS Analysis, but now I want to learn to develop new bioinformatics tools. What will be the best starting point?
Can you clarify if "know how to do NGS analysis" only refers to using existing tools?
Do you have programming experience (in some language) or would that be the first requirement as you embark on "learning to develop new tools"?
I know perl language, R and linux commands.
have a look here http://software-carpentry.org/
Is Software Development a key part of Bioinformatics?
Adding to all the answers and comments, try to find what is lacking in the existing tools and develop your tools accordingly.
Choose a language in which you want to code, and come up with a great idea for a tool.
The most used languages in bioinformatics are R and python but you probably know that already. Here is a comparison of both (a bit outdated but still useful) that can help you choose.
Personally, I prefer to glue stuff together in Bash scripts but I have to admit that Python is often more convenient if premade tools don't exist. For example, reverse_complement() a seq object instead of some combination of rev and tr or whatever..
It sounds like rosalind.info may be up your alley: http://rosalind.info/problems/locations/
Go to GitHub before starting.
I'd start here and do all the exercises, then work out what you want to do from that:
Thanks a lot for your valuable inputs.
Login before adding your answer.
Use of this site constitutes acceptance of our User Agreement and Privacy