Writing your first academic paper

Writing your first academic paper

If you were working in academics (and you were if you were working with Jeff then is at least for the moment) you'll n Eed to write some papers. This are designed to help you get started on writing your first paper since the process can seem a little weird at fi Rst. This are focused on writing your first "first author" paper and not on collaborative papers which involve more Intera Ctions with other groups.

When to write your first academic paper?

As soon as possible! The purpose of Graduate school is (in some order):

• Freedom
• Time to discover new knowledge
• Time to dive deep
• Opportunity for Leadership
• opportunity to make a name for yourself
  • R Packages
  • Papers
  • Blogs
• Get a Job

The first couple of years of graduate school is typically focused on (1) Teaching all the technical skills you need a ND (2) data dumping as much hard-won practical experience from + experienced people into your head as fast as possible.

After that one of your main focuses should is on establishing your own program of the and reputation. Especially for Ph.D. students it can is emphasized enough no one would care about your grades in graduate school BU T everyonewould care about you produced. See for example, Sherri's excellent guide to CV ' s for academic positions.

Your Advisor believes firmly that R packages and blog posts is just as important as papers, but the primary signal to MOS T traditional academic Communities still remains published peer-reviewed papers. Should get started on writing them as soon as can (definitely before you feel comfortable enough to try to writ e one).

Even if you aren ' t going to is in academics, papers is a great to show off so you can (a) identify a useful project , (b) Finish a project, and (c) write well. So the first thing should was asking when you start a project is "what paper are we working on?"

What's an academic paper?

A scientific paper can be distilled to four parts:

1. A set of methodologies
2. A Description of data
3. A Set of results
4. A set of claims

When you (or anyone else) writes a paper The goal are to communicate clearly items 1-3 so that they can justify the set of Claims you is making. Before you can even write down 4 of you has to do 1-3. So, is the where you start when writing a paper.

How does you start a paper?

The first thing you are decide on a problem to work on with your advisor. This can is a problem that your advisor thought of or it can is a problem you is interested in, or a combination of both. Ideally your first project would have the following characteristics:

1. Concrete
2. Solves a scientific problem
3. Gives you a opportunity to learn something new
4. Something you feel ownership of
5. Something you want

Points 4 and 5 can ' t be emphasized enough. I'll try to come and a problem, but if you don't feel like it's your problem it'll make writing T He first paper a total slog. Hopefully we can find an option where is just insanely curious to know the answer at the end, the just has to figure it out and kind of don ' t care about the answer is. That's doesn ' t always happen, but this makes the grind of writing papers go down a lot easier.

Once You has a problem the next step is to actually do the. I ' ll leave this for another guide, but the basic idea is so you want to follow the usual data analytic process:

1. Define the question
2. Get/tidy the data
3. Explore the data
4. Build/borrow a model
5. Perform the analysis
6. Check/critique Results
7. Write things up

The hardest part for the first paper was often knowing where to stop and start writing.

How does know when to start writing?

Sometimes this is a easy question to answer. If you started with a very concrete question at the beginning then once do enough analysis to convince yourself That's the answer to the question. If the answer to the question was interesting/surprising then it was time to stop and write.

If the answer isn ' t interesting/surprising but you started with a concrete question it's also time to stop and write. But things often get more tricky with this type of paper as the most journals when reviewing papers filter for "interest" so s Ometimes a paper without a really "big" result would be harder to publish. This is ok!!  even though it may take longer to publish the paper, it's important to publish even results that aren ' t surprising/ Novel. I would much rather so come to an answer you is comfortable with and we go through a little pain trying to get it p Ublished than you keep pushing until you get a "interesting" result, which may or could not be justifiable.

If you started with a question that wasn ' t so concrete then it gets a little trickier. The basic idea, here, is, you have convinced yourself, which has a result that's worth reporting. Usually this takes the form of between 1 and 5 figures that show a coherent stories that you could explain to someone R field.

For the first paper you'll get a lot of help from me in deciding when to stop. But in the general one thing you should being working on graduate school are your own internal timer that tells you, "OK we hav E-Done enough, time-to-write this ". I found this one of the hardest things to learn in graduate school, and if you were going to stay in academics it was a crit ical skill. There is rarely deadlines for paper writing (unless is submitting to CS conferences) so it'll eventually be You are to start writing. If you don't have a good clock, this can really slow down your ability to get things published and promoted in academics.

One good principle to keep on mind are "the perfect is the enemy of the very good" another one was that a published paper in A respectable journal beats a paper you just never submit because your want to get it into the ' best ' journal.

How does you start writing?

1. Once you has a set of results and is ready-to-start writing up the paper the first thing was not to write. The first thing you should does is create a set of 1-4 publication-quality plots (see Chapter 10here). Show these to Jeff to get confirmation on them before proceeding.
2. Start a project on overleaf and invite the Jeff to join.
3. Write up a stories around the four plots in the simplest language you feel you can get away with, while still reporting all Of the technical details that you can.
4. Go back and add references under only if you have finished the whole first draft.
5. ADD in additional technical detail in the supplementary material if you need it.
6. Write up a reproducible version of your code this returns exactly the same numbers/figures in your paper with no input par Ameters needed.

What is the sections in a paper?

Normally when writing your first paper it'll be is about a statistical method you have developed to analyze a parti Cular set of data, although sometimes it may is an analysis paper.

Keep in mind that most people would read the title of your paper only, a small fraction of those people would read the ABSTR Act, a small fraction of those people would read the introduction, and a small fraction of those people in-read your Whol E paper. So make sure-get to the point quickly!

The sections of a methods paper are:

2. Title: Should is very short, no colons if possible, and state the main result. Example, "A new method for sequencing data so shows how to cure cancer". Here's want to make sure people would read the paper without overselling your results-this is a delicate balance.
4. Abstract: In (ideally) 4-5 sentences explain (a) what problem is solving, (b) Why people should care, (c) Ho W you solved the problem, (d) What is the results and (e) a link to the software you developed implementing the method.
6. Introduction: A More lengthy (1-3 pages) Explanation of the problem is solving, why people should care, and How is solving it. Here's also review what other people has done in the area. The most critical thing was never underestimate how to little people know or care about "what" is working on. It is the your job to explain to them why they should.
8. Methods: This is the meat of a Methods paper. Should state and explain your statistical model, what is the parameters is, how do you chose them, and any strengths or WEA Kenesses of your proposed approach.
10. Comparisons: Compare Your proposed approach to the state of the art methods. I believe in doing this with simulations (where you know the right answer) and data you haven ' t simulated (where do you don ' t Know the right answer). If you can base your simulation on data, even better. I also believe in simulating both the easy case (where your method should is great) and harder cases where your method mig HT be terrible.
12. An exampleanalysis: It was good to having one compelling "use case" laid out of the beginning to the end. Ideally this is the problem so motivated you to create the method in the first place.
14. Conclusions: Summarize what do you do and explain why do you do is important one more time. Provide a link to the software implementing your method and the code for used to perform the analysis.
16. Supplementary information: Sometimes, if there is a lot of technical computational, experimental or statistical Details, you can include a supplement that have all of the details so folks can follow along. As far as possible, try to include the detail in the main text but explained clearly.

The length of the paper would depend a lot in which journal you is targeting. In general the Shorter/more concise the better. But unless is shooting for a really glossy journal (ask Jeff before doing this first) you should try to include the D Etails in the paper itself. This means most papers would be in the 4-15 page range, but with a huge variance.

The main difference with a analysis paper was you present the analysis/results you be doing first and move all of the Met Hods and comparisons to after (or to the supplement).

Working with Jeff

When writing your first paper the usual, we'll do it'll be:

1. You write a draft of the paper
2. Schedule a block of time to re-write the paper with Jeff (usually 2-3 hours).
3. Based on the re-write finish up the new draft
4. Edit the paper
5. Submit

The first draft should is as complete as your think you can make it. Ideally this is something your think you could submit. If you haven ' t written a academic paper before this was almost never ready to submit. You should isn't feel bad about this or let it slow you down. Your goal is to write a draft.

The purpose of the re-write with Jeff is 2-fold. First, we'll get the draft in ' publication Ready ' shape. Don ' t feel bad if this involves a massive re-write. My advisor completely (and I mean completely) reworked the text in my first couple of papers as I was learning. Which brings us to the second purpose of this joint writing session-to Learn how a paper was written and what was think ing about when writing.

The joint writing session would almost never be enough time to tidy up all the details, it'll usually focus on big points . So and you'll need to edit the paper to get all of the details/references ready to go.

Finally, you ' ll submit the paper! This usually involves meeting with Jeff again to go through the manuscript submission process in case there is any files Missing. Unfortunately, most manuscript submission systems is a bit clunky and this can take the up to an hour to get everything enter Ed/submitted. Almost always we'll submit the paper simultaneously to a journal and to a pre-print server like BIORXIV.

The process is actually just getting started

When you submit a paper you'll feel this huge sense of accomplishment and relief. You should!!

But the process was actually only getting started. To get a paper published would also require:

1. Finishing up and distributing software
2. Receiving and responding to referee comments (the target of a future tutorial)
3. re-writing the paper until it gets accepted

Once your paper is accepted the get to celebrate, HOORAY!!!! But you still aren ' t done. The most of the impact from a paper (especially in methods) comes from what's happens after the paper is published. You'll need to maintain the software, respond to questions, fix any typos in the paper (which'll still happen), and Co Ntinue to ' support ' the work do you do. But you'll have the satisfaction of a amazing paper making you want to doing all this extra work ...

Developing a Style

Scientific papers is written in a particular style. The best-of-learn how to read these papers are to go to some journals, regularly publish good papers like Biostatis Tics or genome biology and read a ton of papers.

Learn how to write papers in a very clear and simple style. Whenever you can write in plain 中文版 and make the approach is using understandable and clear. This can (sometimes) make it harder to get your papers into journals. But simple, clear language leads to much higher use/reading/citation/impact of the your work.

I learned by mimicking the styles of other people. Examples of great writers Are:john Storey, Nancy Reid, Shirley Liu, Brad Efron, and Robert may, but there is a lot of OT Her people so can learn from. As with any writing, the more you read, the better a writer you'll be.

I also find the Hemmingway App a good to keep my writing simple.

A couple of other guides, May is useful

1. Guide to Reviewing
2. Guide to Writing Software
3. Guide to sharing data

Writing Your first academic paper