I suggest you that you compose new sequences by concatenating gene1 and gene2 (always in that order) for each of your taxanomic units. Then align the new concatenated sequences and then do a phylogenetic tree by incorporating gamma rate categories, because your new sequences will likely have different rates of evolution. I think all this can be done in popular phylogenetic programs, such as MEGA for example. The first step though, for concatenating sequences is not clear to me if MEGA can do it, but you can do a simple program to do that job. If you need assistance on this just let me know. Best regards.
Well, I should not have said that it was going to be a simple program since I ignore the input format in which you have your sequences. So, let me assume this format and this sequence example:
>RPB.species1
AGCACGGCATCAGC
>EF1.species1
AGCGGCGAGCGAGCGGC
>EF1.species2
GGACTTCGAGCGGCGCA
>RPB.species2
AGAGATTCTCTACC
As you can see I put the sequences not in order, to demonstrate that the program I am going to show you here can deal with that. I selected PHP as the programming language since I am more used to it, but of course it can be translated to any other programming language.
A program that will do the concatenation, for this particular input format of sequences is called concat_sequences.php in the zipped attachment. There I include also the example input file and the output file that is created after you execute that program. The PHP program needs to be execute from a terminal like this:
php concat_sequences.php
please take a look at that file. You will see I assumed a particular name for the input file and for the output file. The latter is called "concatenated_sequences.fna" and will look like this for the particular input in the above example:
>species1.EF1.RPB.
AGCGGCGAGCGAGCGGCAGCACGGCATCAGC
>species2.EF1.RPB.
GGACTTCGAGCGGCGCAAGAGATTCTCTACC
This PHP program is pure PHP, without any dependency. Of course it could have been written much shorter at the expense of installing hundred of dependencies. I prefer independent programs instead.