gchen98 / macs

Automatically exported from code.google.com/p/macs
16 stars 6 forks source link

specifying proportion for splitting command #34

Open abwolf opened 8 years ago

abwolf commented 8 years ago

Hi Gary,

I just wanted to clarify the role of the 'p' value when using -es in macs. In the documentation it describes it as the proportion of chromosomes that move from pop i to pop i+1. So if I have a command macs ... I 2 1 1 ...-es 0.001 1 0.25..., this means that 25% of the chromosomes from pop1 will be moved to pop3?

This is different from how -es functions in ms, where the 'p' value specifies the probability of a chromosome ending up in population i. i.e. -es 0.001 1 0.25 in ms means that a chromosome from pop1 has a 25% probability of remaining in pop1, and a 75% probability of being moved to pop3?

Thank you.

gchen98 commented 8 years ago

yes it should be the reverse of ms based on the source code here:

                 for (NodePtrSet::iterator it=pCoalescedNodes->begin();
                 it!=pCoalescedNodes->end();++it){
                     NodePtr node = *it;
                     if (node->getPopulation()==iSourcePop &&
                         pRandNumGenerator->unifRV()<dProportion){
                         nodesToMigrate.push_back(node);
                     }
                 }

hence the higher the value of p the higher the chance of a migration.

On 8/4/15 4:19 PM, abwolf wrote:

Hi Gary,

I just wanted to clarify the role of the 'p' value when using -es in macs. In the documentation it describes it as the proportion of chromosomes that move from pop i to pop i+1. So if I have a command macs ... I 2 1 1 ...-es 0.001 1 0.25..., this means that 25% of the chromosomes from pop1 will be moved to pop3?

This is different from how -es functions in ms, where the 'p' value specifies the probability of a chromosome ending up in population i. i.e. -es 0.001 1 0.25 in ms means that a chromosome from pop1 has a 25% probability of remaining in pop1, and a 75% probability of being moved to pop3?

Thank you.

— Reply to this email directly or view it on GitHub https://github.com/gchen98/macs/issues/34.