Forum: PhD research proposal - Need current topics
gravatar for David_emir
3.6 years ago by
David_emir360 wrote:

Hi all,

I got an opportunity to pursue my PhD in bioinformatics, I am looking for current trends in bioinfo, need to select one research question and propose it. I wanted to do this before my guide rub his interest on me. i am interested in gene expression patterns with machine learning. Please provide your suggestions and current topics in general.Its an open discussion.


Thanks a ton


research phd forum • 2.6k views
ADD COMMENTlink modified 10 months ago by phdresearchguidance0 • written 3.6 years ago by David_emir360

If you are out of ideas for the research grant, that demands this from you, in my opinion it is not for you. It is cut for students, who already did some work and got caught up in the topic. And to do that you have to put months and months of hard work (mostly reading and testing if those ideas were already tackled).
I would suggest you to go and find phd program with query already stated. Just read on it and justify if it appeals to you

ADD REPLYlink written 3.6 years ago by stolarek.ir600

See also Can We Have A Collection Of Interesting Research Topics For Thesis At Different Levels for an earlier attempt. I'm not very glad about it but I have to disappoint you, biostars is good for asking questions but it's not a good idea-generator, has been tried many times before, never worked out afaik, and likely never will.

ADD REPLYlink written 3.6 years ago by Michael Dondrup46k

It is also very important to state your educational background ( like computer science, bioinformatics etc )

ADD REPLYlink modified 3.6 years ago • written 3.6 years ago by geek_y9.8k
gravatar for John
3.6 years ago by
John12k wrote:

> Doesn't want supervisor's interests to rub off on him.
> Asks random people on the internet for ideas.

Something doesn't add up...

I'm not making fun of you - i'm really not. It's our culture of only getting grants in areas that are seen as sexy that has spurred you to find the most-sexy topic to research. Although it's a stupid approach, it's also completely logical :P I would even find it funny if it wasn't so depressing...

Please sit and think about what challenges you have encountered, and what you would be willing spend your time on to fix, before you submit a grant proposal for Neural-Network-Self-Driving-GPU-Sequencing-Machines-In-The-Cloud.

ADD COMMENTlink modified 3.6 years ago • written 3.6 years ago by John12k

I like the topic,let me add one: Neural-Network-Self-Driving-GPU-Sequencing-Machines-In-The-Cloud-Crispr/Cas9

ADD REPLYlink written 3.6 years ago by Michael Dondrup46k

You forgot this:


ADD REPLYlink written 3.6 years ago by RamRS24k
gravatar for Medhat
3.6 years ago by
Medhat8.4k wrote:

Welcome and I can not give you bullet point answer, but I can help you to think in another way;

The first issue to bear in mind when considering a topic is how viable it is for a PhD project. Generally speaking, most PhD students appear to start their PhDs with over-ambitious projects and find it difficult to focus their initial research question. The key is to ensure that the big topic can be turned into a manageable research question. That is what time and time again supervisors and other successful academics will tell the poor PhD candidate. And yet, how do you actually do that? 

You can read one of those books;

  1. Authoring a PhD: How to Plan, Draft, Write and Finish a Doctoral Thesis or Dissertation (Palgrave Study Guides)
  2. The Unwritten Rules Of Phd Research (Open Up Study Skills)

If you don't have time or need speedy answer;


Lastly good luck :) .


ADD COMMENTlink written 3.6 years ago by Medhat8.4k
gravatar for Benn
3.6 years ago by
Benn7.7k wrote:

I can help you with some ideas, don't know if it interests you.

First idea is to develop an accurate de novo assembler for whole genome (human) data derived from HiSeq X-ten. Preferably one that runs on a desktop computer and is quick. This way we don't need to depend on a reference genome anymore for alignment.

Second idea is to develop a tool that would make nanopore sequencing more accurate (i.e. without the errors).

Just some thoughts...

ADD COMMENTlink written 3.6 years ago by Benn7.7k
Please log in to add an answer.


Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 2140 users visited in the last hour