1
$\begingroup$

I am a bioinformatics student and I was learning the basics of structural equation modeling (SEM).

I was wondering, are there any applications of SEM on bioinformatics? Are there any strong sides for use of structural equation models using data on bioinformatics?

I did a quick search and found almost nothing. While it seemed like a learning algorithm, I can't say I figured it out completely.

Improve this question
$\endgroup$
2
  • 1
    $\begingroup$ I am not sure I understand your question, could you edit to clarify it? What do you mean by "strong sides for bioinformatics"? What do you want to know about bioinformatics? Or are you "just" asking about the algorithm of SEM? BTW here is a paper using SEM. $\endgroup$
    – llrs
    Nov 8, 2017 at 9:22
  • $\begingroup$ I mean, I haven't come across the usage of SEM in bioinformatics related articles. Bu it is widely used in education, questionaries and behavioral studies. In summary, I was wondering whether this method suitable for genomics studies or not. $\endgroup$
    – Ugur Toprak
    Nov 8, 2017 at 11:12

2 Answers 2

Reset to default
1
$\begingroup$

I have came across them in gene regulatory network analysis before. see here and here. However I haven't seen them in other contexts, but that's not to say they don't exist.

Improve this answer
$\endgroup$
1
$\begingroup$

SEM specialises in identifying 'latent variables'.

SEM are extensively used for questionnaire data. In questionnaire data, latent variables are the 'psychological' factors behind given (human) behaviours.

Applying SEM to explicit molecular mechanisms doesn't make much sense unless there is a hidden component in a mechanism that has not been identified. However, even here the SEM will not tell you what the 'latent variable' represents.

Thus I don't believe there is a good application to bioinformatics, outside human phenotypes derived from questionnaire data, which might be mapped on biological variables, but I could be wrong.

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.