I've analysed about 100 patient samples for the genotype of a particular SNP (half from benign ovarian tumours, half from malignant ovarian tumours). The patient data was only given once all samples had been analysed. The SNP associated with a particular histological subtype, and I would like to analyse further samples to determine if this was observed from a sampling error or is representative of the true picture.
I have access to more patient samples and would like to analyse the SNP in this subtype specifically. Are there any studies that have been carried out in this manner i.e. carried out a preliminary analysis while blinded, and then conducted further analysis without blinding? Is it acceptable to do so?
Hi Martin,
I dont see any problem in your design. Since you have found an association between a SNP and a tumor subtype already, you can further test this in more samples also. Your pilot experiment was a blinded one, and if your results on those 100 have statistically significant difference in distribution of the SNP, go ahead testing it in more samples with known subtype. you will find many published studies, which have similar design of study.
Best of luck,
SD
HI Martin,
I agree with Sibnarayan. You can extend your study to further support your initial findings. Make sure to use the very same method for the SNP detection as before.
Good luck!
hello martin,
i would suggest you to increase your sample size since genetic polymorphism studies mostly need a larger sample size to give an authentic result.
so you take more samples and continue your work on the same i hope you will get significant values if there is any such association present and that would also authenticate this preliminary finding of yours but as Pucker suggested use the very same method for SNP detection.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India