questions about ratio.txt about freec HOT 3 OPEN

Hi,
-Inf should be log(0). These must be segments where FREEC predicts copy number of zero.

By visualization, I meant to visualize the results as .png to visually evaluate the amount of noise after the normalization and the quality of FREEC's CNA calls.

from freec.

Hi,
I have a rough idea of your advice.I should visualize ratio.txt to remove the noise and outlier.Later I'll take care of it.

Our research is drawing to a close.We reviewed all the steps and collated them.I am currently working on the first draft of my article.As I review the FREEC, I have a few questions.

1. 0 means Zero copies of DNA in this region predited in _CNVs file. And 0 always corresponds to a loss according to my output.But how to explain it? 0 fragment is lossing from this region? Or too much 0 means noise,and I should filter them, as you
   adviced before?

Part of my _CNVs file:
NC_048218.1 68935104 69748872 1 loss
NC_048218.1 72301368 72327936 0 loss
NC_048218.1 89715216 89736864 0 loss
NC_048218.1 94680480 94689336 7 gain
NC_048218.1 98317344 98352768 1 loss
NC_048218.1 98685360 98714880 0 loss
NC_048218.1 100452624 100478208 0 loss
NC_048218.1 109823256 109882296 0 loss
NC_048218.1 115849272 116669928 3 gain
NC_048218.1 130496112 130513824 0 loss

2. The CNVs FEEEC predicted show two types: loss and gain. I wonder if FREEC writes the normal region into the output -- the copy number does not change compared to the reference genome. If 0 means no change ,why does "loss" appears?

Wish your reply!

from freec.

Hi, I have a rough idea of your advice.I should visualize ratio.txt to remove the noise and outlier.Later I'll take care of it.

Our research is drawing to a close.We reviewed all the steps and collated them.I am currently working on the first draft of my article.As I review the FREEC, I have a few questions.

1. 0 means Zero copies of DNA in this region predited in _CNVs file. And 0 always corresponds to a loss according to my output.But how to explain it? 0 fragment is lossing from this region? Or too much 0 means noise,and I should filter them, as you
   adviced before?

Part of my _CNVs file: NC_048218.1 68935104 69748872 1 loss NC_048218.1 72301368 72327936 0 loss NC_048218.1 89715216 89736864 0 loss NC_048218.1 94680480 94689336 7 gain NC_048218.1 98317344 98352768 1 loss NC_048218.1 98685360 98714880 0 loss NC_048218.1 100452624 100478208 0 loss NC_048218.1 109823256 109882296 0 loss NC_048218.1 115849272 116669928 3 gain NC_048218.1 130496112 130513824 0 loss

2. The CNVs FEEEC predicted show two types: loss and gain. I wonder if FREEC writes the normal region into the output -- the copy number does not change compared to the reference genome. If 0 means no change ,why does "loss" appears?

Wish your reply!

Hi,
I 'm sorry to trouble you. I found that I forgot a truth -- there are 2 chromosomes.

And I missed the meaning of CN. I misunderstood it to the number of copy number region. I always tried to understand CN as how many losses or gains there are. But in reality the two are not equal.

So the anwser is 0 and 1 both represent loss. And CN=2 represents the normal fragments ,which FREEC writes in the ratio.txt.

Thanks to this review, I have to make some adjustments to my research.Best wishes!

from freec.