Analyzing Microarray Data with Confirmatory Factor Analysis
Author Information
Author(s): Crijns Anne PG, Gerbens Frans, Plantinga A Edo D, Meersma Gert Jan, de Jong Steven, Hofstra Robert MW, Vries Elisabeth GE, van der Zee Ate GJ, Bock Geertruida H, te Meerman Gerard J
Primary Institution: University Medical Center Groningen and University of Groningen
Hypothesis
Can Confirmatory Factor Analysis (CFA) effectively identify latent factors related to cisplatin resistance in ovarian cancer cell lines using two-color microarray data?
Conclusion
The study demonstrates that CFA is an effective method for analyzing two-color microarray data when a pre-defined hypothesis is incorporated into the experimental design.
Supporting Evidence
- Two latent factors representing differences in cisplatin resistance were identified.
- 315 genes associated with cisplatin resistance were selected from the identified factors.
- The differential expression of 16 genes was validated with reverse transcription-polymerase chain reaction.
Takeaway
This study shows that scientists can use a special math tool called Confirmatory Factor Analysis to find important genes that help cancer cells resist a drug called cisplatin.
Methodology
The study applied Confirmatory Factor Analysis to two-color microarray data from four ovarian cancer cell lines to identify genes associated with cisplatin resistance.
Limitations
The design ideally requires 'SelfSelf' hybridizations, dye swaps, and independent replications, which may not always be feasible.
Participant Demographics
Four ovarian cancer cell lines with varying degrees of cisplatin resistance.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website