How to write a scientific paper?

One of the Prof. told me the reason why some of his paper is greatly accepted in the community. It is because they are written well in an easy-to-understand manner. Also, they have a value proposition, more on this later.

Sentence Structure

Sentences are the building blocks of the paragraph. It is also a close approximation of the human thought. When we read a sentence, we don't form a meaning after reading the first few words to comprehend the sentence. We have to read the whole sentence and then understand. For example, the sentence "A bird in the hand is worth two in the bush" is neither about the bird nor the bush. For this reason, the reader needs to hold the representation of the earlier words to understand the complete sentence. It is particularly difficult if the sentence structure is too complex, especially for non-native speakers. Therefore, it is important to use a simpler sentence structure. Keep your sentences simple and avoid adding unnecessary clauses and adjuncts. If a sentence is too large break it down into smaller sentences. Don't use jargon just to sound smart. The point is to get the message across, you will have a better chance of doing so with easy-to-digest sentences.

Paragraph Structure

If sentences are building blocks of the paragraph, paragraphs are building blocks of the sections. Each paragraph should contain one single coherent thought or idea. If you are changing the topic, start a new paragraph. The first line of the paragraph is the most important. It tells the reader what this paragraph is all about. The reader decides if it is worth reading by the first line. For example, if I am an experimentalist I might not be interested in the minor theoretical details. One professor once told me that you should be able to understand a paper by reading all the first lines of all the paragraphs. For the result section, state the key findings in the first line of the paragraph. Use the rest of the paragraph to justify that claim. We (I) often tend to give some background, how we did it etc, before actually stating the key findings. But be bold, and start with the key findings with a reference to the figure (if any).

Section structure

Sections are building blocks of the paper. The most common style used in the scientific community is Introduction, Methods, Results, and Discussion (IMRAD) or some variation of it. The introduction should give a brief background of the topic, and state the current gap or question this paper is going to address and why it is important. The methods section is for explaining the scientific protocols, or computational models used to study the aforementioned problem. The Results section is to state the key findings of the paper supported with figures. Lastly, the discussion section should contain a summary, interpretation of the results, limitations of the studies, and possible future directions. One important thing to remember is to never mix the sections, keep them separate. For example, don't show any result in the method section, and don't conclude in the result section. Sometimes it is tempting to give interpretation, just after stating a result. Hold that for Discussion section.

General suggestions

Once, a professor told me that when writing a paper, it's often useful to think like an entrepreneur. Ask yourself, what is the "value proposition" of the paper? Why should anyone on earth read it? For example, if you're a theoretician building a model, write a paragraph in the discussion section about how the claims can be experimentally verified. Also, Propose possible future research directions for a first-year grad student looking for research projects.