Reading for Research

[S.A.M.]

*disclaimer* the opinions expressed in this thread are my own and do not constitute immutable fact. Please consult your advisor before undertaking any exercise.

I thought, since there appears to be a diffidence in the Journal club members, regarding review of the articles, I would put down a few points about the basics of a research article.

This may also be useful for non-science members.


PART ONE: THE BASICS

A journal article can be of of the following types:

1. Original research
2. Review, Perspective or Opinion
3. Methods

A review or perspective generally summarises the available evidence to provide an overview of the known information in a narrow field or offer fresh perspectives on what has already been done. These are very useful for the beginner who wants to get background information about an aspect of e.g. biology, as also for the serious researcher who wants to be up to date or is looking for fresh ideas.

An original research paper is logically divided into the following sections:

1. Title
2. Abstract
3. Introduction
4. Methods
5. Results
6. Discussion
7. Conclusion

The title should be short and informative: what was done, how and by whom

The abstract is a summary of the paper. It contains one line introducing the problem, the objective, methods, results and relevance.

The Introduction should contain a statement of the problem, background or review of the literature, purpose and expected results.

The method should define the subjects, instruments or reagents, procedure and data analysis, including statistics.

The results section reports the results succinctly without interpretation.

The discussion should reflect knowledge and insight describing the interpretations derived from the study.

A methods paper describes a method or review of methods, with validation tools as well as the advantages and disadvantages of each.

 

[draqon]

what happened to one of the most important pieces of a research paper? The Appendix.

 

[Jozen-Bo]

*disclaimer* the opinions expressed in this thread are my own and do not constitute immutable fact. Please consult your advisor before undertaking any exercise.

I thought, since there appears to be a diffidence in the Journal club members, regarding review of the articles, I would put down a few points about the basics of a research article.

This may also be useful for non-science members.


PART ONE: THE BASICS

A journal article can be of of the following types:

1. Original research
2. Review, Perspective or Opinion
3. Methods

A review or perspective generally summarises the available evidence to provide an overview of the known information in a narrow field or offer fresh perspectives on what has already been done. These are very useful for the beginner who wants to get background information about an aspect of e.g. biology, as also for the serious researcher who wants to be up to date or is looking for fresh ideas.

An original research paper is logically divided into the following sections:

1. Title
2. Abstract
3. Introduction
4. Methods
5. Results
6. Discussion
7. Conclusion

The Introduction should contain a statement of the problem, background or review of the literature, purpose and expected results.

The method should define the subjects, intruments or reagents, procedure and data analysis, including statistics.

The results section reports the results succintly without interpretation.

The discussion should reflect knowledge and insight describing the interpretations derived from the study.

Thanks for the format!!!
It will be useful!!


Best Regards,
Jozen-Bo

 

[S.A.M.]

what happened to one of the most important pieces of a research paper? The Appendix.

In journal research articles there are rarely appendices, though supplemental info is often provided.

I will further discuss each part of the research article in detail, to provide an overview describing what it should contain.

This will aid in reading a paper critically and will also provide you with guidelines about writing your own.

Luckily I am writing a paper this week, so you can all benefit from my hands on experience.

PART TWO: ELEMENTS

The elements to scrutinise when reading a paper.

TITLE:

• Does the title tell you what the article is about?
• Does the title tell you what, whom and how?
• Does the title make you want to read the article?

ABSTRACT:

-Does the abstract contain

• purpose
• method
• results
• conclusion
• relevance

-After reading the abstract could you grasp the essential points in the study?

INTRODUCTION:

Do the authors

• state the problem?
• provide current literature in the background?
• identify a gap in the knowledge?
• state the purpose of their study?
• define a hypothesis?

METHOD:

This should include

• subject description
• sample selection/exclusion criteria
• sample size
• control group present or absent
• instruments clearly identified
• detailed procedure of experiments
• ease of replication wrt reagents, methods, skills
• internal validation of methods
• description of data analysis
• level of significance
• appropriate statistical analysis

RESULTS:

Look for:

• summary of data accurate and descriptive
• statistical significance of results
• clinical or biological significance of results

DISCUSSION:

• was the hypothesis accepted or rejected?
• are there weaknesses in the experiments or design as identified by the author?
• is there literature to back up the findings
• are the results applicable in a clinical setting?
• are there suggestions for future research?

CONCLUSION:

• do the authors briefly restate the results?
• do the conclusions follow accurately from teh findings?

 

[Exhumed]

I'm really curious what the reading speed of others is when reading these things. Personally I read them slowest of all things I read.

Thanks for the format!!!
It will be useful!!


Best Regards,
Jozen-Bo

Seconded. : )

 

[S.A.M.]

If you keep a copy of the elements with you and start following the format it will get easier.

Just take a notepad and jot down the points covered in the list above. You'll be surprised how much easier and faster you comprehend.

My advice is to read a paper in this order

1. Title
also read the names, locations; the first author is usually the project leader (usually because sometimes there are nasty politics at play), the last author is the PI.

2. Abstract
this will give a swift grasp of the objective results and relevance

3. Introduction
whats the problem and why they did the study

4. Discussion.
This usually covers all the findings and their significance
List it
Finding A - significance
Finding B-significance

5. Results and Methods.

Select one figure, check the method used. Summarise the result in one sentence, then look for gaps, based on the list above

On a good day, I can go through 50 papers.

The trick is to decide before you read the paper what you want to get out of it.

 

[Exhumed]

50? O_O Is that normal?

 

[S.A.M.]

50? O_O Is that normal?

Depends, you have to wade through a lot of stuff to search for what you want. A usual critique (half page, single space) requires at least 30 good references. A more elaborate review (3 pages double spaced) should have at least 50. When I prepare for lab meetings or journal clubs, I usually read not only the paper under review but also a few other related papers, it gives an edge in discussions.

Working in clinical means you need to optimise every protocol, find literature using all permutations and combinations of a procedure before you decide which one is most suitable for you.

e.g. if I were doing a western to identify levels of PGC-1alpha in muscle tissues, I would need to research the best antibody, the best dilutions, find a protocol that has been validated, check for quirks and errors, decide the best homogenisation buffer, the best isolation protocol, the best dilution for primary and secondary antibody.

[Usually after preparing the tissue samples, I make different concentrations of the proteins using a pooled sample (5ug, 10ug, 15ug, 20ug) then different dilutions of primary and secondary antibody, run several of them on one gel, repeat 2x and select the best.]

Then I go back and see what other peoples results looked like, if mine is the same or different etc. Takes about two months to optimize a protocol, with all the permutations and combinations.

Then back to the literature for interpretation, find data that supports your findings see what other people had to say about it, etc.

The worst is when you are doing something new (also the best in a way, since nothing you do is wrong)

Takes a lot of reading, and if you have more than one project going on, takes longer. I did a lot of optimisations in the first two years along with lit review, so now I can sit back and just run the stuff.

 

[Exhumed]

Cool, thanks for the continued valuable info.

You seem like a pro's pro. Compared to the faculty at my school anyway. Maybe I underestimate them, but going by the rate some of them go through student papers I can't imagine it.

 

[S.A.M.]

Cool, thanks for the continued valuable info.

You seem like a pro's pro. Compared to the faculty at my school anyway. Maybe I underestimate them, but going by the rate some of them go through student papers I can't imagine it.

I'm not surprised, I run into problems because I'm too slow; well duh, no one else goes through the whole process.

But its fun, sometimes I run concurrent gels with different levels of milk protein in the blocking buffer, sometimes I play around with the TBST, sometimes with the washing times. I also like to read on the principles of the method, what each stage of the process involves. I like to know the nitty gritty.

But students papers make dull reading

 

[Exhumed]

I'm not surprised, I run into problems because I'm too slow; well duh, no one else goes through the whole process.

Heh

But its fun, sometimes I run concurrent gels with different levels of milk protein in the blocking buffer, sometimes I play around with the TBST, sometimes with the washing times. I also like to read on the principles of the method, what each stage of the process involves. I like to know the nitty gritty.

I (and probably everyone) like to play around too. My school should offer a bio lab for just that purpose. That is when I learn the most, rather than just following a protocol step by step, when often only knowing basic concepts about the materials being used.

But students papers make dull reading

Of course! I agree that student papers shouldn't be assigned.

 

[S.A.M.]

Heh

I (and probably everyone) like to play around too. My school should offer a bio lab for just that purpose. That is when I learn the most, rather than just following a protocol step by step, when often only knowing basic concepts about the materials being used.
Of course! I agree that student papers shouldn't be assigned.

Now you know what I meant about the funding :m:

 

[BenTheMan]

Depends, you have to wade through a lot of stuff to search for what you want. A usual critique (half page, single space) requires at least 30 good references. A more elaborate review (3 pages double spaced) should have at least 50.

And you can't forget to cite all of your friends

 

[BenTheMan]

For the Physicist

SAM asked for some input, and I can say that these things vary considerably between fields. Some friends of mine and I meet about two-three times a week just to read SCIENCE papers, usually about monkeys and ants and such, and there are some pretty big differences between the fields. And even across the disciplines of physics there are some pretty big differences.

Also, I can only really talk about theoretical physics, as it is the only type of paper that I have ever written.

For us, I would class the papers as (in increasing length)
-Letter
-Procedings
-Report
-Article
-Review.

In phyiscs (theory, at least), the adherence to the strict scientific method outline (a la what SAM described in post # 1) is more or less thrown out the window. There is usually an introduction where you cite as many people as you can, laying out more or less the basic idea. You can also anticipate results here. The next section is typically a background section---for example, if I were writing a paper about a new way to derive the radiation from black holes, I might spend a few pages outlining the original calculation of Hawking.

The next several sections are generally where you present your calculations, in as much detail as is necessary. In general, having a lot of equations is a bad thing, but somethimes unavoidable.

The last section is a Conclusion, where you try to point out as many related ideas so that you can get more people to cite you in their Introductions.

This format is very loose, and can be completely ignored with few consequences. It also depends on the TYPE of paper you write. more on this later.

 

[S.A.M.]

And you can't forget to cite all of your friends

Oh heavens yes! Not only that, you must remember to acknowledge as many people as possible without sounding like an Oscar winner. After all, you may need them again!

PART THREE: (A) THE TITLE

You might think, whats the big deal about a title?

For starters, the title is the first impression of your work. If your title is not clear, descriptive and enticing, people may not even read the abstract, let alone the article.

If its not succinct it may not accurately portray your work.

If it is loaded with technical jargon, readers may miss the point you are trying to make.

If it does not contain a recognised authority in the list of names that follow, the approach towards it will be slightly more skeptical and critical

The title, in short, conveys your ability to accurately portray your work and ability to communicate it while simultaneously according it a level of reliance based on peer recognition.

Here is an example

---------------------------------------​

PGC-1{alpha} protects skeletal muscle from atrophy by suppressing FoxO3 action and atrophy-specific gene transcription
Marco Sandri*,{dagger}, Jiandie Lin*,{ddagger},§, Christoph Handschin*,{ddagger}, Wenli Yang*,{ddagger}, Zoltan P. Arany{ddagger}, Stewart H. Lecker*,¶, Alfred L. Goldberg*,||, and Bruce M. Spiegelman*,{ddagger},**

---------------------------------------​

From reading this title, you know that its a gene expression study, related to muscle metabolism and addresses prevention or treatment of atrophy of muscle tissue. And that Bruce Spiegelman, who is De Man is your GOD.

To reiterate, the first author is usually the project leader and the last, the Principal Investigator, aka the one who does the grunt work to collect the money for your research. The PI is also usually the corresponding author and the one to whom all requests for reprints, donations of reagents, primers, antibodies and complaints about your stolen ideas in the artcle may be addressed.

Take Home:
A good title should give insight into what (was done), whom (it was done to) and how (it was done).


Remember, people who peruse the literature have precious little time to skim through it. Even if your study is well designed and well executed with interesting and significant findings, if your title is poor many people will miss reading it.

If the title is enticing, it will motivate people to read the Abstract <coming soon>

 

[BenTheMan]

Take Home:
A good title should give insight into what (was done), whom (it was done to) and how (it was done).

...without alerting the readers as to how boring the work actually IS

 

[CharonZ]

Actually there are also differences between journals and not only between disciplines. On the other hand, the differences between disciplines may not be so large either. Granted, theoretical physicists write very different papers, however many experimental physiscs paper have a similar setup as described by Sam. In the end, the journal decides.
Of course, all the types of papers that Ben describes, are also found in various Biojournals (e.g. letters and proceedings, though the latter are only sometimes reviewed).

Likewise, depending on journal and the type of paper the structure may be different. There is always some kind of introduction, but results and discussion can be mixed and conclusions might be omitted. Methods can be very short and integrated into the results&discussion section (in letters).

...without alerting the readers as to how boring the work actually IS

Absolutely!

 

[S.A.M.]

...without alerting the readers as to how boring the work actually IS

Absolutely!

Or how often it was rejected and re-submitted!

 

[CharonZ]

Depends on the quality of the title of course

 

[S.A.M.]

Before writing your own research paper, you should address three questions:

1. What is the central message of the paper?

2. What is the new contribution you want to make?

3. What do you want to convince people about?

Do a thorough literature search; even as you write teh article to show your awareness of the current research.

Make bullet points of interest and build into paragraphs before summarising.

Now moving on to the next part of the article:

PART THREE: (B) THE ABSTRACT

The abstract is found at the beginning of the article.
It is a summary that reduces the entire article to a single paragraph.

Sometimes the abstract is published by itself, e.g. in Pubmed search engines, in article indices; and often the article can be searched in librarary databases by keywords present in the abstract. This makes the abstract a critical element of the paper. Different journals, competitions and conferences have different requirements for the number of words which can be used in an abstract and sometimes it is a challenge to get all your points across in a limited space. Usually special characters are a bad idea in an abstract, as they may not publish properly everywhere.

Sometimes, for conference submissions, a figure is included in the abstract.

The abstract contains an introduction, the purpose and methods of the study, outlines the major results and presents a brief summary of the conclusions or teh clinical relevance.

Basically the abstract tells the reader what to expect in the article.

To save time, many readers read only the abstract, coming back to the article if details or indepth information is required.
It is a time efficient way to determine if the article contains information relevant to the readers search.

Hence, the abstract also functions as a screening device, which can be critical when sifting through a lot of literature.

One more point I forgot to mention is that even when articles are not available due to restricted access, the abstract of the article usually is and should help you to decide if you need that particular article for review. (saves $$$)

Example of an abstract:

1: J Biol Chem. 2008 Apr 14 [Epub ahead of print]

Signaling and crosstalk by C5a and UDP in macrophages selectively use PLCbeta 3
to regulate intracellular free calcium.

Roach TI, Rebres RA, Fraser ID, Decamp DL, Mean-Lin K, Sternweis PC, Simon MI,
Seaman WE.

Dept of Immunology, Northern California Institute for Research & Education, San
Francisco, CA 94121.

Studies in fibroblasts, neurons, and platelets have demonstrated the integration
of signals from different G-protein coupled receptors (GPCRs) in raising
intracellular free Ca2+. To study signal integration in macrophages, we screened
RAW264.7 cells and bone marrow-derived macrophages (BMDM) for their Ca2+ response
to GPCR ligands. We found a synergistic response to complement component 5a (C5a)
in combination with uridine 5'-diphosphate (UDP), platelet activating factor
(PAF) or lysophosphatidic acid (LPA). The C5a response was Gai-dependent, while
the UDP, PAF, and LPA responses were Gaqdependent. Synergy between C5a and UDP,
mediated by the C5a and P2Y6 receptors, required dual receptor occupancy, and
affected the initial release of Ca2+ from intracellular stores as well as
sustained Ca2+ levels. C5a and UDP synergized in generating
inositol-1,4,5-trisphosphate, suggesting synergy in activating phospholipase C
(PLC) ss. Macrophages expressed transcripts for three PLCss isoforms (PLCss2,
PLCss3, and PLCss4), but GPCR ligands selectively used these isoforms in Ca2+
signaling. C5a predominantly used PLCss3, while UDP used PLCss3 but also PLCss4.
Neither ligand required PLCss2. Synergy between C5a and UDP likewise depended
primarily on PLCss3. Importantly, the Ca2+ signaling deficiency observed in
PLCss3-deficient BMDM was reversed by reconstitution with PLCss3. Neither PI-3
kinase nor PKC was required for synergy. In contrast to Ca2+, PI3-kinase
activation by C5a was inhibited by UDP, as was macropinocytosis, which depends on
PI3- kinase. PLCss3 may thus provide a selective target for inhibiting Ca2+
responses to mediators of inflammation, including C5a, UDP, PAF, and LPA.
​

Source