search

galosh / profillic-hmmer

Modified versions of HMMER 3 programs to implement interoperability with galosh profiles (as generated by profillic).
http://galosh.github.com/profillic-hmmer
GNU General Public License v3.0
2 stars 3 forks source link

prolific, DynamicProgramming.hpp, argument incompatibility #7

Open tedholzman opened 12 years ago

tedholzman commented 12 years ago

in the DynamicProgramming.hpp file, on line 8087 there is a min() function call. The arguments are of different types that C++ finds incompatible. Changing the line to read:

m_sequence_count = ( ( sequence_count == 0 ) ? sequences->size() : min( (size_t) sequence_count, sequences->size() ) );

i.e. adding (size_t) cast, repairs it.

pedlefsen commented 12 years ago

Hi Ted. I wonder if replying to this will work? My vague belief is that casts in c++ should preferably be like: static_cast(value).. Paul

On Fri, Nov 4, 2011 at 5:14 PM, Ted Holzman reply@reply.github.com wrote:

in the DynamicProgramming.hpp file, on line 8087 there is a min() function call.  The arguments are of different types that C++ finds incompatible.  Changing the line to read:

m_sequence_count = ( ( sequence_count == 0 ) ? sequences->size() : min( (size_t) sequence_count, sequences->size() ) );

i.e. adding (size_t)  cast, repairs it.

Reply to this email directly or view it on GitHub: https://github.com/galosh/profillic-hmmer/issues/7

tedholzman commented 12 years ago

On 11/4/11 5:25 PM, Paul T. Edlefsen wrote:

Hi Ted. I wonder if replying to this will work? My vague belief is that casts in c++ should preferably be like: static_cast(value).. Paul

On Fri, Nov 4, 2011 at 5:14 PM, Ted Holzman reply@reply.github.com wrote:

in the DynamicProgramming.hpp file, on line 8087 there is a min() function call. The arguments are of different types that C++ finds incompatible. Changing the line to read:

m_sequence_count = ( ( sequence_count == 0 ) ? sequences->size() : min( (size_t) sequence_count, sequences->size() ) );

i.e. adding (size_t) cast, repairs it.

Reply to this email directly or view it on GitHub: https://github.com/galosh/profillic-hmmer/issues/7

Reply to this email directly or view it on GitHub: https://github.com/galosh/profillic-hmmer/issues/7#issuecomment-2638084 Yes. I think that ought to work. I think the incompatibility is in the "const" storage of one versus the stack storage of the other.

               -Ted
pedlefsen commented 11 years ago

---------- Forwarded message ---------- From: Diem, Kurt kdiem@fhcrc.org Date: Mon, Aug 13, 2012 at 3:43 PM Subject: different herpes project To: "Paul T. Edlefsen" pedlefsen@gmail.com

**

Paul,

I know you are busy but there is another herpes project Larry would like you to consider. Surprisingly, at this time the number of circulating strains of HSV2 in the US is poorly understood and more investigation could impact vaccine design. The AiCuris company has been testing a new drug for herpes treatment. We have been doing traditional sequencing to check for drug resistance in two genes, UL5 and UL52. I have made simple neighbor joining trees and to my eyes there may be more than one strain present but I believe you have more phylogenetic tools and skills than I do. From this data can we say anything about the minimum number of clades circulating in the US and is there evidence of recombination among clades? See attached for NJ tree files(actually UL5 is a ML tree). If you would like the sequences themselves or more information, let me know.

Kurt 206-667-3117