Hi there
I am a bit confused about how the default parameters in the default.cfg file are described (and chosen):
# Weight for each hint source
# Values have to be >= 0
P 1
E 20
C 1
M 1
# Required fraction of supported introns or supported start/stop-codons for a transcript
# Values have to be in [0,1]
intron_support 1.0
stasto_support 2
# Allowed difference for each feature
# Values have to be in [0,1]
e_1 0.1
e_2 0.5
e_3 0.05
e_4 0.18
For the middle part, it says "values have to be in [0,1] which I read as if they should be in the interval 0 to 1 (as fractions normally are). But then why is stasto_support 2?
Furthermore, there are four e values in the last part, but in the text there is mention of six features, and in the example in the text (readme), also six are shown. It is also confusing that the parameters in the example are so different than the default, without any explanation.
It should also be noted that even if it says in the config files that the values have to be fractions, it clearly works anyways. The old default for e_3 and e_4 were 25 and 10, respectively, and it is still possible to run the latest version of Tsebra with those parameters.
Lastly, the default parameters have changed somewhat with the different versions, but I cannot find any information as to why the changes were made. The idea of having a default (to me) is that it should work well in most cases - but if the defaults are changing that much, it becomes difficult to understand how they were chosen. Looking at the original paper is not helpful when the values used there were so different. I also tried to compare the versions but the config files did not come up as files that were changed.
Is it possible to clarify any of this?
Sincerely,
Sjannie
Hi there
I am a bit confused about how the default parameters in the default.cfg file are described (and chosen):
For the middle part, it says "values have to be in [0,1] which I read as if they should be in the interval 0 to 1 (as fractions normally are). But then why is stasto_support 2?
Furthermore, there are four e values in the last part, but in the text there is mention of six features, and in the example in the text (readme), also six are shown. It is also confusing that the parameters in the example are so different than the default, without any explanation.
It should also be noted that even if it says in the config files that the values have to be fractions, it clearly works anyways. The old default for e_3 and e_4 were 25 and 10, respectively, and it is still possible to run the latest version of Tsebra with those parameters.
Lastly, the default parameters have changed somewhat with the different versions, but I cannot find any information as to why the changes were made. The idea of having a default (to me) is that it should work well in most cases - but if the defaults are changing that much, it becomes difficult to understand how they were chosen. Looking at the original paper is not helpful when the values used there were so different. I also tried to compare the versions but the config files did not come up as files that were changed.
Is it possible to clarify any of this?
Sincerely,
Sjannie