Entering edit mode
3.7 years ago
Lucas
▴
20
Hi
For NIPT analysis there are some free software packages like wisecondor, WisecondorX, Sanefalcon and Defrag; but all these tools need to be trained correctly to give us an accurate result, as I tested some other trained models from other labs and they didn't work correctly on my samples, BUT what about the commercially developed NIPT diagnosis systems?
Imagine an Asian lab sends its samples to an European NIPT center for analyzing, so they use their own model on it, but the answer is usually accurate?!
What is their strategy?
Did they train different models for each country or zone in the world?
For example the trained models for the European people will not get us an accurate result for analyzing Asians.
Hii Lucas, Have you used Sanefalcon for fetal fraction detection?
To be honest, I have an impression that "training" is done to take the level of variability in your sequencing method into account. It has nothing to do with the population. Asian samples, sequenced by the same people with the same protocol in the same lab, will have results of same accuracy that European ones.