Abstract
    The abstract is a quick summary of what is in the paper.  It generally includes a sentence or two about what the basic question is, a number of sentences describing the techniques used and results obtained, and a summary sentence of why these results are important.  The abstract is most useful for those already familiar with the research area, and provides a quick way to judge if the paper is important to the research question involved.

Introduction
    The introduction sets the stage for the rest of the paper.  This generally includes a somewhat extended section on the general topic of the paper, and includes many references to previous work.  It is designed to put the actual research results into perspective, since no science is done in a vacuum.  The introduction places the coming material into the context of what is already known about the topic at hand, and may contain a quick summary of the results at the end of the introduction.

Materials and Methods
    The materials and methods section describes exactly how the experiments were conducted.  Well, that's the theory, anyway.  Oftentimes procedures are referenced (and redirected again, etc.) and summarized, or performed "according to instructions in the kit."  This section is vitally important to anyone interested in duplicating the results, as it lists appropriate concentrations, specific antibodies, buffers, etc. that make all the difference in the world when trying to obtain results.  It is also important for interpreting results reported in the paper, as experimental conditions and the precise experimental design can change a clear, unambiguous result into a questionable muddle.  Unfortunately, Science and Nature, two very prestigious journals, have minimal materials and methods, making these articles somewhat difficult to evaluate fully.  If you are designing experiments, this is a critical section of the paper.

Results
    This is the core of the paper.  Here is where all of the data is presented with details, explainations, figures, tables, and most of the critical information for interpreting the data presented (the materials and methods section can also be essential).  Results can be as simple as a western blot showing the tissue distribution of a protein, to a multi-panel immunofluorescence staining figure with accompanying intensity graph and standard errors.  The experiments performed will determine the form of the figures, and the type of figure generated from different experiments can vary significantly even using similar techniques.  The results section will also generally include the researcher's logic and interpretation of the results used to design following experiments that amplify or confirm earlier results.  When critically reading a paper, this section will provide the most important elements needed to judge the rest of the paper.  Brilliant prose in the introduction or a fantastic model in the discussion don't mean much if the actual data is shaky or worse, simply bad.  Conversely, well designed experiments with clear results and a well designed experiemental approach are priceless.  A good results section will allow an experienced researcher to essentially write the introduction and discussion sessions themselves.

Discussion
    The discussion section re-examines the data from the results section and attempts to put it into a larger framework.  This can include developing a model of how a particular protein functions or may simply describe how these experiments are similar or different to those performed previously by others.  In some ways this is the most interesting section to read, as it draws the data from the results and explains how it agrees or disagrees with current theory.  It can also be very useful for putting complicated experimental results into a simplified framework.  The danger, however, is that if the results are ambiguous or misleading or simply misinterpreted, the discussion is worthless.  Good results are critical for a good paper.

References
    Science is never done in a vacuum; it is always based on previous work.  The references section lists the citations used in the paper and is invaluable for those looking for more information about the research area.  A good references section includes both reviews in the area of the paper, as well as relevant, significant findings that bear upon the experimental design and interpretation of the results.
 

Critical Reading of the Scientific Literature

    Critically reading the current literature is an essential skill for those contemplating a career in biology.  It is not necessarily a skill that comes naturally and, like any skill, becomes easier and more natural with practice.  In some ways, critical reading is a combination of two different skills.  First, it is the ability to examine and interpret the data without bias, as if collected directly from the experiment, tearing apart weaknesses and deciding if the presented material actually shows what the authors claim.  Does this picture really show good colocalization, or are the pictures so overexposed as to make any comparisons meaningless?  Did the scientists use appropriate statistics (or any statistics) to evaluate the significance of the results?  Sometimes this means really examining the experimental method in detail to decide if the result is real, or whether the conditions used skew the result in one direction or another.

    The second skill is deciding if these are the most appropriate experiments to prove what the authors want to claim.  Is there a better, more direct way to obtain the same conclusion?  Can a particular technique be used in a given situation?  This is a difficult aspect of critical reading for young scientists to develop, as they do not necessarily know what caveats are associated with various techniques or what alternative methods could be used to obtain similar results.  This skill also includes evaluating whether all of the appropriate controls are included to show that the technique is working as expected.

    Critical reading of the literature is an art.  It is very easy to dismiss results as being too vague, not significant, open to interpretation, etc.  However, it is far easier to destroy a manuscript than it is to create one.  Some techniques are simply less quantitative or less definitive than others, yet still provide the best current methods for addressing a particular issue.  Many scientific advances are based on incremental improvements in technology and techniques that allow better data to be collected.  (Behavioral tests in neurobiology are notorious for their crude nature, yet everyone uses the same restricted set of experiments because they are the only ones generally available.)  For any set of results, it is important to understand how the various experiments fit together to substantiate each other.  If all of the major findings are based on a single experiment of questionable value, then the entire paper is weak.  If, on the other hand, a somewhat weak experiment is used to confirm a finding that is suggested by several other experiments, the paper can still be a strong one.  The only way to become good at critically reading the literature is reading broadly and interpreting the data in the light of what others in the same field are doing.