Testing heterotachy model

[Homap]

Hello,

I have run iqtree2 on my concatenated alignment and have obtained the best set of models for the partitions identified by iqtree2 as such:

  charpartition mymodels =
    SYM+I+R3: p1_atpA_p13_pafI,
    GTR+F+I+R4: p2_atpB_p21_psaA_p25_psbA_p27_psbC_p28_psbD,
    TVM+F+I+R3: p3_atpE_p4_atpF_p20_petL_p33_psbJ_p37_psbT_p38_psbZ_p41_rpl14_p42_rpl16_p43_rpl2_p44_rpl20_p47_rpl5_p48_rps11_p50_rps14_p51_rps18_p56_rps7,
    GTR+F+I+R3: p5_atpH_p15_pbf1_p19_petG_p23_psaC_p26_psbB_p40_rbcL_p49_rps12_p59_tufA,
    GTR+F+I+R3: p6_atpI,
    TIM3+F+I+R4: p7_ccsA_p63_rpoC1a,
    TVM+F+I+R4: p8_cemA_p9_chlB,
    TVM+F+I+G4: p10_chlL,
    TVM+F+I+R4: p11_chlN_p12_clpP_p61_rpoBa_p62_rpoBb_p64_rpoC1b,
    TVM+F+I+R3: p14_pafII_p39_psb30_p52_rps19_p53_rps2_p55_rps4_p57_rps8,
    TPM3u+F+I+R3: p16_petA_p24_psaJ_p31_psbH_p45_rpl23,
    TIM3+F+I+G4: p17_petB_p22_psaB,
    TIM2+F+I+G4: p18_petD,
    TPM3u+F+I+R2: p29_psbE_p30_psbF_p32_psbI_p34_psbK_p35_psbL_p36_psbM_p46_rpl36,
    TVM+F+I+G4: p54_rps3_p58_rps9,
    TIM3+F+I+R3: p60_rpoA;

I have a total of 55 species in my alignment of size 50981 bases. Previous studies have indicated that a heterotachy model might be more suitable for chloroplast DNA. So, I wanted to test that. Does this make sense to change this file into:

  charpartition mymodels =
    SYM+I+H2: p1_atpA_p13_pafI,
    GTR+F+I+H2: p2_atpB_p21_psaA_p25_psbA_p27_psbC_p28_psbD,
    TVM+F+I+H2: p3_atpE_p4_atpF_p20_petL_p33_psbJ_p37_psbT_p38_psbZ_p41_rpl14_p42_rpl16_p43_rpl2_p44_rpl20_p47_rpl5_p48_rps11_p50_rps14_p51_rps18_p56_rps7,
    GTR+F+I+H2: p5_atpH_p15_pbf1_p19_petG_p23_psaC_p26_psbB_p40_rbcL_p49_rps12_p59_tufA,
    GTR+F+I+H2: p6_atpI,
    TIM3+F+I+H2: p7_ccsA_p63_rpoC1a,
    TVM+F+I+H2: p8_cemA_p9_chlB,
    TVM+F+I+H2: p10_chlL,
    TVM+F+I+H2: p11_chlN_p12_clpP_p61_rpoBa_p62_rpoBb_p64_rpoC1b,
    TVM+F+I+H2: p14_pafII_p39_psb30_p52_rps19_p53_rps2_p55_rps4_p57_rps8,
    TPM3u+F+I+H2: p16_petA_p24_psaJ_p31_psbH_p45_rpl23,
    TIM3+F+I+H2: p17_petB_p22_psaB,
    TIM2+F+I+H2: p18_petD,
    TPM3u+F+I+H2: p29_psbE_p30_psbF_p32_psbI_p34_psbK_p35_psbL_p36_psbM_p46_rpl36,
    TVM+F+I+H2: p54_rps3_p58_rps9,
    TIM3+F+I+H2: p60_rpoA;

and use this as the partition file to obtain a new ML tree? Then to compare which model fit best, to compare them by their BIC score.

Thanks very much for your help,
Homa