Tips for paper writing

I. Format

Example Manuscripts


  1. Title and Authors

  2. Abstract and Keywords

  3. Introduction

  4. Results

  5. Conclusion and Discussion

  6. Methods

Font and Line Space

  1. Check Font: Times New Roman, 11pt or 10.5pt

  2. Check Line: double space (except for References)

  3. No indent for the first sentence of a new paragraph.


  • (1) Background - (2) Results - (3) Conclusions


  • Replace with the key words we want to be searched. If number of key words is limited, we may not repeat with title only if this keyword is very important.

II. Writing Rules


  1. Check the writing of specific terms in the whole paper:

    • noncoding, cross-validation, P-value, V.S. (need to be capitalized), HCC/liver cancer

  2. Check if the Abreviations are consistent in the whole paper. Check if full name is used when it is mentioned first time. Occasionally, we may use full time several times in the beginning of Introduction, Methods and Suppl. Info., because they are separated documents when published.)


Grammarly ( is very helpful. It has web version, desktop (MAC/Windows) version and chrome/firefox/safari plugin.

  1. Check plural or single.

  2. Check if Past tense is used when we describe results (e.g., "We sequenced 50 samples ...").

  3. Check if Present tense is used when we make a claim or conclusion (e.g., usually in the last sentence of a paragraph, "This result suggests that …" ).

Figure and Figure Legend

Figure and Figure legend sometimes are more IMPORTANT than main text, because many senior readers read figures first.

  1. Check if Figure Legends contain sufficiently detailed descriptions. (We got this suggestion from a previous reviewer who rejected our paper.)

  2. Check if the colors are easy to see. (We got this suggestion from a previous reviewer who rejected our paper.)

Main Text

  1. Check first and last sentence in the whole paper. To make it easy to read, in every paragraph, first and last sentences are important. Usually, first sentence introduces and summarizes the whole paragraph; last sentence concludes the whole paragraph (e.g. "This result suggests that ...").

  2. Check voice: 1. In Results, we prefer active voice, such as "We analyzed the data …", not "The data were analyzed …". 2. In Methods, we may use passive voice sometimes , such as " The difference was tested by a paired t test.".

  3. Check sentence patterns (Useful Sentence Patterns):

    1. We performed RT-qPCR (RNA-seq …) for … → XX was validated (seqenced ...) by RT-qPCR.

    2. We built a Random Forest model … → We used/developped a Random Forest model to …

    3. We assayed/analyzed … datasets.

    4. We curated/collected … datasets from … OR 100 datasets were curated/collected from ...

    5. XX was significantly enriched in the noncoding sequences (t test P-valeue: 0.002).

    6. The enrichment is/was calcuated with RNApromo (ref.). The P-value cutoff is/was 0.001; the xx is/was 0.5; the xx is/was 0.8. (depends on if you are talking to the audience about a plot or describe a result you have done.)


  1. Check every result to see if it needs statistical test: Was a test used? Which one? What is the P-value?

  2. Check every enrichment to see if it was statistically determined to be enriched: What test was used? Which one? What is the P-value?

III. 网友的写作建议


  • 两步走:1、写出尽可能多的内容,不要担心语法、拼写、组织;2、修改文稿变得更好。(参考海明威的写作方式)

  • 当你写作时,不要编辑!写作和编辑使用不同的半脑,交叉这两件事不容易。

  • 每个结果和相应的讨论一起写,以免之后忘记结果的解释和意义。再写方法,最后写介绍(基于结果和讨论)。

  • 用照片做类比:

    • Introduction:Why was the picture taken?

    • Results:What do you see?

    • Conclusion and Discussion:What does it mean?

    • Methods:How was it taken?


  • 阅读投稿期刊的『Instructions for authors』,了解期刊期望的行文风格。

  • 介绍部分说明本研究的理由,不是文献综述。讨论部分不用『可能』,如果不确定结论,做更多的实验。

  • 结果部分不使用『could』、『can』(受中文影响),减少不确定性。

  • 不使用『etc』、『and so on』,这些词说明作者懒。

  • 『observed』很弱很被动,改用:assess, calculate, detect, estimate, evalulate, examine...

  • 不用『obviously』,读者不一定容易理解。

  • 结果部分避免『significantly』,改为报告p-value。讨论部分可用。