[Uncle Skeleton]

I'm quite fed up with office for writing scientific documents. Word formats away my formatting, it's a PITA to make sub- and super-scripts, Excel can't draw error bars, etc ad infinitum. There must be a better solution. what is it? where are all those fresh unemployed Mol-Bio-Comp-Sci students to write me an Office for scientists?

[Logic]

I completly agree with you. But unfortunatly I have no idea about any app that is better. Just thought I'd let you know that you aren't alone in this predicament

[calumr]

You haven't heard of LaTeX? I use iTexMac, which is the nicest Cocoa front-end out there at the moment IMO.

http://itexmac.sourceforge.net/

TeX (and LaTeX) are used by almost everyone in Math/CS/Physics to format documents, and the author Donald Knuth is one of the most respected in CS. He offered $2 for the first bug in TeX, $4 for the next, then $8 etc. It got up to about $512 I think, and then no more bugs have really been found since. The thing is, he has so much respect that no-one wants to cash in the cheques, they just frame them and put them on their wall.

Could you imagine if Billy G. offered an exponentially increasing reward for bugs in MS Word?

[Hozie]

Yep, LaTeX is the way to go. Can't believe you haven't heard of it if you're a Sci major... Anyway, you have to install the TeX binaries first and then get yourself a front-end for the package. I suggest going with Gerben Weirda's iInstaller for the binaries:

ftp://ftp.nluug.nl/pub/comp/macosx/volumes/ii2/II2.dmg


Just run it and follow the instructions. You need the TeX foundation and TeX programs to run TeX.

Then, for the front-end, I suggest the highly respected TeXShop (versiontracker has it). There are other GUI's out there, but this one's the nicest IMHO. Finally, you have to learn the syntax. There are plenty of tutorials out there, and you can check back once everything's installed...

[Powaqqatsi]

OmniGraffle

[Leibowitzn]

If you're going to use latex I'd recomend purchasing two things: a copy of the latex manual and a copy of omni graffel. You'll find both are life savers.

[wallinbl]

Originally posted by Uncle Skeleton:
Mol-Bio-Comp-Sci

That's an interesting combo. I majored in Molecular Biology, and now I am a programmer.

[m@]

Use Latex, it isn't as hard as you think.

I use TeXShop, Illustrator 10 for diagrams (as editable pdf's) and a little macro for Excel to make latex tables out of a spreadsheet (as it is alot easier to put data into excell, check, and reformat, and THEN put it into a LaTeX table).

The hardest part is to stop fideling with the presentation (this should be done at the end) and concentrate on the content, which is the whole idea behind LaTex. Believe me am tyring to write up my PhD at the moment, and it looks too nice at present!

For me the main advantage is being able to touch-typing equations, and the high quality output.

As for instructions, there is a web downloadable PDF file called "The Not So Short Introduction to Latex 2e", by Tobias Oetiker, which comes with the tex files for the documents so has all the examples you will need. This is the only document I use on a regular basis.

There is a good FAQ at http://www.tex.ac.uk/.

[kman42]

Originally posted by wallinbl:
That's an interesting combo. I majored in Molecular Biology, and now I am a programmer.

Are you wrting Mol Bio apps or did you completely change fields? Did you teach yourself to program or go back to school?

kman

[Brazuca]

TeX is definetelly the way to go. i was looking for something that could handle my dissertation, and I found out about TeX, and in one night I had all my equations on it. They look beautiful and I don't have to deal with Word's crap.

There is a slight learning curve, but once you get the basic principles you'll find that its easy to learn/do the specific things you want to do. There is also a very helfpul users group out there.

Start here

[pimephalis]

Since no one's tackled this one, I'll deal with the plotting/graphing issue.

There are a couple of commercial solutions available, but they are pretty pricey if your department/grant isn't going to foot the bill. If you're looking for a free solution, I'd suggest looking seriously at gnuplot and R. They're both command line by nature, and thus there is a bit of a learning curve, but they are extremely powerful and can be automated to graph things repetitively.

As an anecdote, I wanted to create a 3x3 matrix of scatterplots with their respective regression lines and confidence envelopes. Doing that it Prism would have taken more than a day to make sure that everything was imported, scaled, aligned and generally ordered. After 15 minutes of typing in my text editor, I had a routine for R that would not only create all of the plots to my specifications but would also perform the statistical analysis. Hard to beat that for speed.



p.s. I love LaTeX as well.

[Jerommeke]

Indeed. Office is one big pain in the ass. However, it does draw error bars, how **** it might be!

[Uncle Skeleton]

office does draw error bars? how?

[spiney]

Double click your series and select the "y error bars" tab.

[Marsico]

If Focault is a scientist . . .

Honestly though, I find that TextEdit and Nisus Writer Express (the new OSX beta version) are both excellent apps for doing basic text entry.

Super/subscipt options can be easily accessed in Texed Edit by hitting Format -- Font -- Baseline ---

Anything such as page layout I think is best suited to non-word-processor applications. In fact, when writing, I really hate even having to contemplate anything that has to do with how my typing is going to look when I print it out.

If I need to make a table of contents, place graphics, set margins, choose a pretty font, or fit the text to a specific style I do it in Adobe Indesign.

So my suggestion is to ditch office -- Text Edit has everything you probably need -- and for the hard stuff buy a copy of InDesign.

InDesign can also make faaar nicer graphs than are possible in Excel.

Marsico

[Hozie]

Originally posted by m@:
Use Latex, it isn't as hard as you think.

I use TeXShop, Illustrator 10 for diagrams (as editable pdf's) and a little macro for Excel to make latex tables out of a spreadsheet (as it is alot easier to put data into excell, check, and reformat, and THEN put it into a LaTeX table).

The hardest part is to stop fideling with the presentation (this should be done at the end) and concentrate on the content, which is the whole idea behind LaTex. Believe me am tyring to write up my PhD at the moment, and it looks too nice at present!

For me the main advantage is being able to touch-typing equations, and the high quality output.

As for instructions, there is a web downloadable PDF file called "The Not So Short Introduction to Latex 2e", by Tobias Oetiker, which comes with the tex files for the documents so has all the examples you will need. This is the only document I use on a regular basis.

There is a good FAQ at http://www.tex.ac.uk/.

Dang, I use EXACTLY the same tools, including that Excel macro. Plus, I used them on my M.Sc thesis as well... Spooky!

[krove]

How sweet would it be if Apple (or any other great Cocoa developer; OmniGroup, you listening?) wrote a scientific graphing/plotting program capable of generating Aqua-quality graphs of various types (bar, line, scatter, etc) complete with modern graphics capabilities (transparency, etc). Add on top of that excellent feature control of the graph itself and easy exporting. I'd easily pay whatever they asked.

Why are graphing/plotting programs so hard to come by? And when you do find one, you find out rather quickly its limitations and how much it really sucks.

[awaspaas]

Originally posted by krove:
How sweet would it be if Apple (or any other great Cocoa developer; OmniGroup, you listening?) wrote a scientific graphing/plotting program capable of generating Aqua-quality graphs of various types (bar, line, scatter, etc) complete with modern graphics capabilities (transparency, etc). Add on top of that excellent feature control of the graph itself and easy exporting. I'd easily pay whatever they asked.

Why are graphing/plotting programs so hard to come by? And when you do find one, you find out rather quickly its limitations and how much it really sucks.

All Apple would need to do would be to add some more power to the graphing function they use in Keynote.

[sgrup]

my vote goes for LaTeX as well - i use TeXShop personally

s

[xtal]

Another vote for TeX. I am a TeXShop fan, but I've recently been fiddling with iTeXMac. I only starting using TeX after a suggestion from this forum, and that was about six months ago. I'm hooked! I use OpenOffice for short notes and letters, but anything longer (papers, reports, research proposals) are all done using LaTeX.

Oh, and I'm a biochemist studying cancer by way of molecular modelling and protein structure.

[alien]

I used LaTeX back in the '90s. It's brilliant for math. Haven't had much use for it the last years, but I have looked at TeXShop, and it's so nice...

There's a website for TeX and Mac
here.

[mineralomac]

If its proper scientific graphs and data analysis you want, then the best advice you will ever hear is:

Igor Pro

www.wavemetrics.com

Cheers

Richard

"Igor is not a program to Richard, it's a religion" - Ben Grguric, my Ph.D. office mate.

[mmurray]

I am a TeXShop user myself but you might like to look at Scientific Word and related software.

http://www.mackichan.com/


I have never used this - has anyone here
tried it ?

Michael

[kooBi]

Originally posted by Uncle Skeleton:
I'm quite fed up with office for writing scientific documents. Word formats away my formatting, it's a PITA to make sub- and super-scripts, Excel can't draw error bars, etc ad infinitum. There must be a better solution. what is it? where are all those fresh unemployed Mol-Bio-Comp-Sci students to write me an Office for scientists?

A few things to make Word more bearable:

Under Tools --> AutoCorrect:

Select the "AutoFormat As You Type" pane: deselect all of the "Apply as you type" items (this will stop some of Word's 'helpful' changes).

Subscript superscript issue: I assume that you know about the key combinations:

Command+Equalskey --> subscript
Command+Shift+Equalskey --> superscript

And yeah, Excel can draw error bars, although you are best off calculating SE/SD yourself in cells adjacent to your mean values, and inputting those SE/SD into the error bar data.

My major gripes with Office: M$, user dictionary functionality, language support, graph/figure/table handling (better stop there)

[riverfreak]

No doubt about it, Office bites.

I'm a geneticist and molecular biologist, but I now spend most of my time writing code for comparative genomics projects.

My tools: XEmacs for text editing, LaTex for manuscripts. Occasionally, I'll use Word for manuscripts with lots of others (for tracking edits).

Graphing: R and gnuplot, and custom scripts in perl using GD.

I second the motion for a powerful cocoa graphicing app. That would be sweet - especially if it had serious statistics and was scriptable for repeat analyses. Can't handle using the damn wizard in office to remake graphs when data changes.

[Ozmodiar]

Originally posted by mineralomac:
If its proper scientific graphs and data analysis you want, then the best advice you will ever hear is:

Igor Pro

www.wavemetrics.com

Cheers

Richard

"Igor is not a program to Richard, it's a religion" - Ben Grguric, my Ph.D. office mate.

In my Experimental Neurobiology lab, my professor wants us to use SigmaPlot (which is, of course, Windows-only). He said we can make our charts and graphs in Excel if we want, but neatness counts and he wants them to look as if they came out of SigmaPlot.

I've been messing around with SP in the lab and have found it is both easier to use than Excel and presents nicer-looking data. Igor Pro looks nice but unfortunately I don't have this kind of money. Are the Tex apps easy to get acquainted with and, in the same token, do they produce nice charts and graphs?

[voodoo]

Why not use the R? It is free and powerful like S+ which is a $500 pro app. We use it extensively (R or S+) in our biology department!

Highly recommended.

http://cran.r-project.org/

[Simon]

Originally posted by Uncle Skeleton:
There must be a better solution. what is it? where are all those fresh unemployed Mol-Bio-Comp-Sci students to write me an Office for scientists?

Well, I don't know about Mol-Bio, but in Physics where I am everything is LaTeX.

I'd get teTeX (the package for TeX, LaTeX, etc.) and TexShop as a GUI front end. This web site tells you all about it: http://darkwing.uoregon.edu/~koch/texshop/texshop.html

Forget Word. That's for babies.

[dagaz]

Originally posted by Uncle Skeleton:
Word formats away my formatting, it's a PITA to make sub- and super-scripts

Look I don't want to sound like I'm sucking up to a Micro$oft product (but I actually think Word is pretty damn good) but here's a few tips. The keyboard shortcut for subscript is command-= (i.e. apple + equal sign) and the keyboard shortcut for superscript is command-shift-= . Two other things I recommend are Word > Preferences > General > untick the box for Include formatted text in clipboard and then go into Tools > Autocorrect ... and play around in there until you have things set up the way you want them. Its not perfect but at least you have the following benefits: integrates with the OSX clipboard and other OSX applications and file compatability with most computer users. That's my 2c worth.

[JKT]

Originally posted by dagaz:
Look I don't want to sound like I'm sucking up to a Micro$oft product (but I actually think Word is pretty damn good) but here's a few tips. The keyboard shortcut for subscript is command-= (i.e. apple + equal sign) and the keyboard shortcut for superscript is command-shift-= . Two other things I recommend are Word > Preferences > General > untick the box for Include formatted text in clipboard and then go into Tools > Autocorrect ... and play around in there until you have things set up the way you want them. Its not perfect but at least you have the following benefits: integrates with the OSX clipboard and other OSX applications and file compatability with most computer users. That's my 2c worth.

Try using it to write a 200+ page scientific thesis and you will understand why people end up loathing it so much. The single biggest mistake of my life so far by a loooooooong way.

FWIW, I'd also recommend trying out TypeIt4Me. It is a systemwide equivalent of Word's autocorrect, only far better and much more useful. It is invaluable if you are constantly typing complex names such as isopropyl-[beta]-thiogalactopyranoside in your text - just give it a shortcut e.g. iptg and when you type that, it will auto-expand to the full word. Auto-correction is a pain when you don't want it to happen, but tremendous when you do!

[Arkham_c]

There was a really good article in LaTex recently over at the O'Reilly MacDevCenter. Here's the URL:

http://www.macdevcenter.com/pub/a/ma.../03/latex.html

Hope that is helpful to all you scientists out there.

Unfortunately, my experience has been that outside of academia, if your document isn't in Word format, nobody can read it

[Simon]

Originally posted by Arkham_c:
Unfortunately, my experience has been that outside of academia, if your document isn't in Word format, nobody can read it

Give them a PDF - TeXShop does pdfTeX so it's just one click away. They can read and print that.

[wataru]

One major hurdle for me using TeX for general things is support for non-European languages. Luckily, that issue is easily solved with this ptex package in conjunction with TeXShop. Complete instructions for installing and using it are included with TeXShop.

[kman42]

How do journals handle your LaTeX documents? A lot of bio journals want them in MSWord. I think Nature and Science are the same, but not positive.

kman

[barney ntd]

Almost all mathematics, logic and computer science journals only accept latex. Elsevier (for example) has its own latex .sty file which formats your paper to match their journal look. In my experience (within the "mathematical sciences") latex is universal, and anyone trying to use word (or anything else) is told to go away and rewrite it.

This is probably why word's mathematical formatting is so poor: professional mathematicians don't use it, so there's no real demand for improvements. In fact, once you get used to latex, it's much easier to type $\int_0^\alpha f(x)\,dx$ than to find all the symbols in some window and fiddle with menus until the positioning looks right.

For those that love latex but are forced to use word (or any other wysiwyg formatting system) I recommend Equation Editor. You type in latex mathematics, and it produces a pdf picture which you can drop into your document.

Barney.

[Neo.cmg]

Originally posted by kman42:
How do journals handle your LaTeX documents? A lot of bio journals want them in MSWord. I think Nature and Science are the same, but not positive.

kman

I've personally never run into any problems publishing my papers among several scientific/engineering journals using TeX. I just hand them the .pdf output and they're happy as can be. Some publications even have templates that you can drop into your one of TeX's directories and at a click of a button reformat things exactly as the publisher wants them.

By the way, I'm an aerospace engineer and have used TeX with Gerben Wierda's teTeX distrubution and Richard Koch's TeXShop for text entry. All graphs are created using gnuplot and sketches/drawings are done using OmniGraffle.

Neo.cmg

[kman42]

As far as Mo Bio journals that accept TeX:

http://www.lecb.ncifcrf.gov/~toms/latex.html

Not totally up-to-date, but seems like a reasonable list.

I guess the important point is that you shouldn't plan on publishing in Nature

kman

[Anand]

The real question is what other word processor can you use to write grants with?

[mudmonkey]

Originally posted by JKT:
Try using it to write a 200+ page scientific thesis and you will understand why people end up loathing it so much. The single biggest mistake of my life so far by a loooooooong way.

Exactly. I did my M.Sc. thesis in Word. It got to over 100MB in filesize and eventually, I hit the bug of, "you inserted another figure. I will corrupt the file, but, generate no errors." I lost a week's worth of work. It took me about 30 hours to get it all back, and from that point on, I had to conserve figures, as, I had hit some limit in Word (on both the Mac and Windows versions). This was in Office 98 way back in 99.

Texshop, bibdesk, gnuplot, GMT, etc. are the tools for my Ph.D. dissertation. Word is used only when some wank sends me a .doc file.

[foamy]

If you are working in Mol Bio, then stick with Office and shell out the $$$ for a decent graphing program. If you need fancy equations, then try something else.

I did my 200 page dissertation in Word and yes, it was a PITA, but I only plan on writing one dissertation. The suggestions above will help a lot. I basically turn everything off possible in Word, customize my toolbar for super/sub scripts and create my own dictionary that I add to religiously. Getting images to stay put is still a harrowing experience, especially when you move one, and it screws up your entire dissertation.

Also the Journals I have published in request .doc or .rtf files for the text and .tif, .psd,.eps etc for artwork.

Grants. If you are applying to NIH for funding, then you're stuck with their horrible .doc and .rtf forms that are truly a PITA.

Anyway, if you didn't know how to do superscripts or set error bars using Office, then I wouldn't count out Office just yet...

[kman42]

Since I have all the scientists attention, I thought I would ask a non-related question in this thread. Anyone have a good bioinformatics website for OSX? Preferably one with some forums for discussing packages, installation, problems, etc.

I found this site for an easy package for installing bioinformatics software:
http://www.ebioinformatics.org/

thanks,
kman

[voodoo]

Originally posted by foamy:
If you are working in Mol Bio, then stick with Office and shell out the $$$ for a decent graphing program.

I'm in mol bio. Why would I shell out for a graphing program when I have the R for free? What kind of graphs do you mean? I'd rather use text edit in combination with XPress myself. But I am fortunate enough to know my way around XPress so that may not apply to everyone.

[voodoo]

Oh and as for a good Tex for biologists, have you tried BioTex?

[wataru]

Originally posted by voodoo:
Oh and as for a good Tex for biologists, have you tried BioTex?

What's the difference?

[voodoo]

Originally posted by wataru:
What's the difference?

http://www.biotex.com/

It was my attempt at humor.

However there is a BioTeX project here http://homepages.uni-tuebingen.de/beitz/tse.html

[wataru]

What a difference just one capitalized letter makes!

[spiney]

I guess even Ecology takes LaTeX documents. Shame on Nature.