for both RosA and RosB, find better model proteins.

[danielhhaft]

Please compare

HHH2222284.1 to your current RosA protein
and
HDL8274934.1 to your currrent RosB protein.

It seems both RosA and RosB are rather ancient, and have regions of corrupted sequence. Probably frameshifted out and then back again.

tpg|HHH2222284.1| TPA: MFS transporter [Yersinia enterocolitica]
MTDRSETELPPSVNTQPFDNTKVKRTSFSILGAISVSHLLNDMIQSLILAIYPLLQAEFSLSFAQIGLIT
LTYQLTASLLQPLIGLYTDKHPQPYSLPIGMGFTLSGILLLAVATTFPVVLLAAALVGTGSSVFHPESSR
VARMASGGRHGMAQSIFQVGGNFGNALGPLLAAILIAPYGKGNVGWFSLAALLAIVVLLQVSKWYQQQQR
ATYGKVVKVSSAKILPKKTVISALVILMVLIFSKYFYLTSISSYYTFYLMHRFGVSVQNAQIHLFVFLFA
VAAGTIIGGPLGDRIGRKYVIWGSILGVAPFTLILPYVSLYWTGVLTVIIGLILASAFSAILVYAQELIP
GKVGMVSGLFFGFAFGMGGLGAAVLGYVADLTSIELVYQICAFLPLLGIITVFLPNIEDK

tpg|HDL8274934.1| TPA: Kef family K(+) transporter [Yersinia enterocolitica]
MHHSTPLITTIVGGLVLAFLLGSLAHRLRISPLVGYLAAGVLAGPFTPGFVADTSLAPELAEIGVILLMF
GVGLHFSLKDLLAVKAIAIPGAVAQIAVATLLGMGLSHLLGWDLMTGFVFGLCLSTASTVVLLRALEERQ
LIDSQRGQIAIGWLIVEDLAMVLTLVLLPAFAGVMGNETTSLSQLFTELAITIGKVIAFITLMIVVGRRL
VPWILAKTASTGSRELFTLAVLVLALGIAYGAVGLFDVSFALGAFFAGMVLNESELSHRAAQDTLPLRDA
FAVLFFVSVGMLFDPMILLREPLAVLASLAIIIFGKSAAAFILVRMFGHSKRTALTISVSLAQIGEFAFI
LAGLGISLGLMSEHGRNLVLAGAILSIMLNPLLFTLLDRYLAKNETMEDLILEEAVEEEKQIPVNLCNHA
LLVGYGRVGSLLGAKLHAEGIPLVVIENSRPRVEALREQGINAVLGNAASADIMSLARLDCARWLLLTIP
NGYEAGEIVASARIKRPDLEIIARAHYDDEVVYISDRGANQVVMGEREIANSMLNMLKIETLTEEDKRPL
CPI

A search of public databases will show that these sequences are far more typical for their families.

Daniel Haft
Staff Scientist
NCBI