HI,

Step 1- After performing MSA - Mafft and cleaning alignment using trimal.

Step 2- I concatenated all the *.fa files into out.phy, additionally generated partitions.txt

catfasta2phyml.pl  *.fas > out.phy 2> partitions.txt

Total no. of partitions - 1080, does my partitions file is in the right format. 

I don't know, exactly which model to be assigned. How do I choose the model for each partition ? 

head  partitions.txt

    OG0015287.mt.r.fa = 1-431
    OG0015288.mt.r.fa = 432-1222
    OG0015290.mt.r.fa = 1223-1594
    OG0015291.mt.r.fa = 1595-1724
    OG0015292.mt.r.fa = 1725-2050
    OG0015294.mt.r.fa = 2051-2215
    OG0015295.mt.r.fa = 2216-2629
    OG0015296.mt.r.fa = 2630-2984
    OG0015298.mt.r.fa = 2985-3441
    OG0015299.mt.r.fa = 3442-4362
    OG0015300.mt.r.fa = 4363-4729
    OG0015301.mt.r.fa = 4730-4841
    OG0015303.mt.r.fa = 4842-5597

I ran the iqtree2 with out providing partitions.txt

The iqtree2.log file, says "9 sequences failed"

66959 parsimony-informative, 58943 singleton sites, 675786 constant sites
                  Gap/Ambiguity  Composition  p-value
Analyzing sequences: done in 0.000985321 secs using 99.66% CPU
   1  A.v     12.30%    failed      0.00%
   2  C.l      1.25%      failed      0.00%
   3  E.d     17.75%    failed      0.00%
   4  E.s     31.46%    failed      0.00%
   5  L.f      20.00%    failed      0.00%
   6  L.g     20.18%    failed      0.00%
   7  L.li     22.04%    failed      0.00%
   8  L.t      4.30%     failed      0.00%
   9  P.a     7.02%      failed      0.54%
  10  P.s     2.50%    `passed`     15.69%
****  TOTAL              13.88%  9 sequences failed composition chi2 test (p-value<5%; df=19)
NOTE: minimal branch length is reduced to 0.000000124737 for long alignment
Checking for duplicate sequences: done in 0.00445477 secs using 1716% CPU

Suggestions appreciated.