Other Books about scientific computing

[quantknight]

Hello, can someone suggest me some introductory, or basic books on scientific computing(computational physics or computational biology) ? I m planning to study masters in it and I want to grasp the basics of it before i commit.

 

[DrClaude]

I suggest that you start by acquiring a good grasp of numerical methods. Numerical Recipes is a good book for that.

 

[quantknight]

I suggest that you start by acquiring a good grasp of numerical methods. Numerical Recipes is a good book for that.

thank you.. is the basic science concepts required for scientific computing?? i have a computer science undergrad.

 

[quantknight]

I suggest that you start by acquiring a good grasp of numerical methods. Numerical Recipes is a good book for that.

hi What i meant was, i have to prepare for the scientific computing ? or for the subjects physics and biology?

 

[DrClaude]

I'm not sure I understand your question. Computational physics (or biology) is a very broad label, that can be applied to many different fields of inquiry. Coming from CS, I think that you should first get familiar with numerical methods (if you are not already) of the kind that are found in Numerical Recipes. As for physics or biology, I think you will have to wait until you have a specific problem to solve, and try to understand the particular science involved.

 

[onoturtle]

I can't speak for physics, but I agree computational biology is very broad. Like say the methods used in computational neuroscience I imagine is heavy on differential equations to simulate the dynamics of neurons firing and such, while computational genomics aka bioinformatics, the area I work in, uses completely different methods (variants of string matching/edit distance algorithms, statistics, machine learning) where numerical methods may not even be the best place to start. OP, you need to narrow down the scope of your question. If you haven't already, learn the basics of physics and biology from a survey course or even a popular science book and pick a couple areas to narrow it down.

 

[Aufbauwerk 2045]

I am not a biologist, but just in case this helps you at all, one of the Perl books I studied some time ago was Beginning Perl for Bioinformatics by Tisdall.

Apparently Perl is very convenient for processing DNA information, at least to the extent it's represented by character strings, and this book contains examples. I do not know if Perl is still widely used for that purpose.

As for computational physics, here are a couple of links.

http://www.cmth.ph.ic.ac.uk/people/a.mackinnon/Lectures/compphys/

http://www.physics.umd.edu/courses/CourseWare/EssentialMathematica/

 

[quantknight]

I can't speak for physics, but I agree computational biology is very broad. Like say the methods used in computational neuroscience I imagine is heavy on differential equations to simulate the dynamics of neurons firing and such, while computational genomics aka bioinformatics, the area I work in, uses completely different methods (variants of string matching/edit distance algorithms, statistics, machine learning) where numerical methods may not even be the best place to start. OP, you need to narrow down the scope of your question. If you haven't already, learn the basics of physics and biology from a survey course or even a popular science book and pick a couple areas to narrow it down.

thank you. I have already started studying the basics for the science subjects. But I am in a confusion, to what extent I should learn "basics" in science books. I cannot be sure whether I should learn upto college level or just up to school level. below is the list of the books I'm learning basics,
physics-conceptual physics, fundamentals of physics by H&R,
chemistry-introductory chemistry by zumadahl,
maths- cambridge igcse books by hugh neil and douglas
biology-biology essentials by marielle hoeffnagels.

So what is the basic level of science i have to learn if I would go for computational geophysics or bioinformatics or computational nueroscience?

Say if I am interested in nueroscience, do I need a bachelors degree in biology or nueroscience? or i can start learning with a computer science degree? As of what I have learned through the basics, I like gravity, electromagnetism and quantum concepts in physics. For biology I would say, bioinformatics, neurosciences and bio-physics.

 

[quantknight]

I'm not sure I understand your question. Computational physics (or biology) is a very broad label, that can be applied to many different fields of inquiry. Coming from CS, I think that you should first get familiar with numerical methods (if you are not already) of the kind that are found in Numerical Recipes. As for physics or biology, I think you will have to wait until you have a specific problem to solve, and try to understand the particular science involved.

thank you

 

[quantknight]

I am not a biologist, but just in case this helps you at all, one of the Perl books I studied some time ago was Beginning Perl for Bioinformatics by Tisdall.

Apparently Perl is very convenient for processing DNA information, at least to the extent it's represented by character strings, and this book contains examples. I do not know if Perl is still widely used for that purpose.

As for computational physics, here are a couple of links.

http://www.cmth.ph.ic.ac.uk/people/a.mackinnon/Lectures/compphys/

http://www.physics.umd.edu/courses/CourseWare/EssentialMathematica/

thank you.

 

[onoturtle]

A list of online resources with some commentary for computational biology: http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1003662

It is a huge list. Read the descriptions, pick out what sounds interesting to you and check them out. You probably shouldn't worry too much about the prereqs since you're just exploring potential interests.

So what is the basic level of science i have to learn if I would go for computational geophysics or bioinformatics or computational nueroscience?

To study those topics in college? Or to get a job in the area? For both, it depends. Bioinformatics courses can be aimed at computer science majors (so you can probably only assume high school exposure to biology) and be more focused on the details of the algorithms. Or can be aimed at biologists (thus assuming just basic computing skills) and be more focused at how to use bioinformatics tools to solve your research problems. And of course there are degree programs for bioinformatics nowadays that try to strike a balance. I think all routes are valid. Working in academia I meet people skewed to one side but also people who are quite balanced.

As for getting a job in the field, you probably need an MS or PhD. I can't think of any bioinformaticians with just a BS. I can't speak for getting a job in industry, but for the academic job market, life sciences, like a lot of other academic areas, is probably over producing PhDs so jobs are highly competitive and you could be a postdoc for a long, long time. I guess that gives you long time to develop both computational and biology skills, but at some point you'll want a decent paycheck and retirement savings!

For my personal road to becoming a bioinformatician, I majored in computer science for BS and PhD. I took a biology course in high school equivalent to the first year college course for life science majors (the textbook used was Campbell Biology) and that was the start of my interest in genetics. Ultimately I picked CS though. In college, I took a genetics course for non-majors, studied a paper on protein folding prediction for an artificial intelligence course (my first exposure to computational biology). Then in grad school I took a bioinformatics seminar (mostly aimed at CS students), which lead me to decide to do my dissertation in that area. That's it for my bio course background. Beyond that has just been from reading papers for my PhD dissertation and my current job as a junior researcher in bioinformatics. It would be nice for myself to look over the above link and go through some additional basics, but... lack of time!

 

[quantknight]

A list of online resources with some commentary for computational biology: http://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1003662

It is a huge list. Read the descriptions, pick out what sounds interesting to you and check them out. You probably shouldn't worry too much about the prereqs since you're just exploring potential interests.

Thank you

To study those topics in college? Or to get a job in the area?

I meant to get into a master's program, what are the basic prerequisites(for both biology and physics)? college level or high school level knowledge is needed?

As for getting a job in the field, you probably need an MS or PhD. I can't think of any bioinformaticians with just a BS. I can't speak for getting a job in industry, but for the academic job market, life sciences, like a lot of other academic areas, is probably over producing PhDs so jobs are highly competitive and you could be a postdoc for a long, long time. I guess that gives you long time to develop both computational and biology skills, but at some point you'll want a decent paycheck and retirement savings!

This is worrying me. I would like to get into a decent paying job whether it is into research or industry. I cannot afford to be jobless or paid less after completing it, as I have to repay my debts. So do I have any chance of getting into a good job after completing masters? And in which computational fields the job prospects are healthy? (like computational neurosciences, computational geophysics, or machine learning). So that I can correlate both decent paying field and my interest, and I would plan to find a program.

 

[quantknight]

sorry, i just wanted to clarify before I make any decision, it would be really helpful if someone replies the below question.

So do I have any chance of getting into a good job after completing masters? And in which computational fields the job prospects are healthy? (like computational neurosciences, computational geophysics, or machine learning

 

[onoturtle]

The machine learning/data science market in industry seems good right now (just not in my city). I don't know about anything else. I think it would be better if you made a separate thread regarding the health of the job market and whatever other career advice you're interested in at the career advice section. That should lead to more eyes to discuss your concerns.

 

[hilbert2]

Protein folding dynamics is a field of computational biochemistry that will probably advance a lot when computers get faster and able to do quantum mechanical simulations of large molecules (this will help with cancer research and the development of new antibiotics and other medications). http://www.faculty.virginia.edu/CompMat/mse524/articles/ARPC-Protein-Folding.pdf

It's difficult to do that kind of work just because of the money you earn from it, you actually need to have an ambition to do good research. But of course everything must be planned in a way that is affordable, I admit.

 

[quantknight]

The machine learning/data science market in industry seems good right now (just not in my city). I don't know about anything else. I think it would be better if you made a separate thread regarding the health of the job market and whatever other career advice you're interested in at the career advice section. That should lead to more eyes to discuss your concerns.

thank you.

 

[quantknight]

Protein folding dynamics is a field of computational biochemistry that will probably advance a lot when computers get faster and able to do quantum mechanical simulations of large molecules (this will help with cancer research and the development of new antibiotics and other medications). http://www.faculty.virginia.edu/CompMat/mse524/articles/ARPC-Protein-Folding.pdf

It's difficult to do that kind of work just because of the money you earn from it, you actually need to have an ambition to do good research. But of course everything must be planned in a way that is affordable, I admit.

thank you.