HMM PD00143
Back to List of HMMs
Symbol: CC1_MTM_MTMR14_CR
Name: MTMR14 conserved regions
Why built the HMM?
We built the profile because the boundary of phosphatase domain defined by Pfam profile is not the same as the one reported in crystal structure study [1] and SMART does not have the profile for the myotubularin phosphatase domain. The crystal structure shows the phosphatase domain ranges from 205-586 in human MTMR2 (Figure 6A in [1]). The Pfam profile Myotub-related (PF06602) predicted the range from 192-529, which lacks the C-terminal ~45 aa.
We PSI-BLASTed the region from 190 to 600 of human MTMR2. The region is the phosphatase domain determined in crystal structure study with flanking regions [1]:
>Hsap_MTMR2:205-586 id=HsapP048_AA CC1:CC1:Myotubularin:MTMR1 [Homo sapiens] GWKLYDPLLEYRRQGIPNESWRITKINERYELCDTYPALLVVPANIPDEELKRVASFRSRGRIPVLSWIHPESQATITRCSQPMVGVSGKRSKEDEKYLQAIMDSNAQSHKIFIFDARPSVNAVANKAKGGGYESEDAYQNAELVFLDIHNIHVMRESLRKLKEIVYPNIEETHWLSNLESTHWLEHIKLILAGALRIADKVESGKTSVVVHSSDGWDRTAQLTSLAMLMLDGYYRTIRGFEVLVEKEWLSFGHRFQLRVGHGDKNHADADRSPVFLQFIDCVWQMTRQFPTAFEFNEYFLITILDHLYSCLFGTFLCNSEQQRGKENLPKRTVSLWSYINSQLEDFTNPLYGSYSNHVLYPVASMRHLELWVGYYIRWNPR
We first PSI-BLASTed the region against RefSeq database via NCBI BLAST server. We found 5000+ hits in 1st round, which is difficult to look into. We then PSI-BLASTed the region against RefSeq database via NCBI BLAST server. The coverage of human myotubularins:
- MTM1, MTMR1, MTMR2: ~190-600, 100%
- MTMR3: ~214-595 (human, rat, mouse); MTMR4: ~225-588 (human, mouse), ~213-598 (xenopus)
- MTMR5: ~317-535 (human), ~317-579 (mouse); MTMR13 (not in swissprot?)
- MTMR6: ~195-595 (human, mouse); MTMR7: 192-592 (human), 203-593 (mouse); 192-593 (human, chicken)
- MTMR9: 192-588 (human, mouse), 204-592 (cow)
- MTMR10 (MTMRA): 214-532 (human, mouse, xenopus, zebra fish); MTMR11 (MTMRB): 214-561 (human, mouse); MTMR12 (MTMRC): 202-556 (human), 225-556 (mouse, rat, zebra fish),
- MTMR13 (MTMRD): 223-581 (human), 345-581 (xenopus)
- MTMR14: no hit
We just built the HMM followed the general guidance. By using the protein phosphatases of the nine genomes, we found the HMM profile can detect the phosphatase domains of all subfamilies with good boundaries except MTMR14. The HMM has a length of 391 aa, all subfamilies except MTMR14 has the domains of > 380 aa long. Because the HMM does not capture MTMR14, we archive the HMM in a separate page CC1_MTM_1_PD.
The CC1_MTM_1_PD HMM profile hit a region of 60 aa in human MTMR14. Because there is not structure of MTMR14, we looked into the conserved region of MTMR14 and investigated whether the region can align with CC1_MTM_1_PD.
How was the HMM built?
Conserved region of MTMR14
We PSI-BLASTed human MTMR14 full sequence against NR database via NCBI BLAST server. Eyeballing the sequences, the query hit MTMR3 and other myotubularins. To build a HMM for full sequence MTMR14, we chose MTMR14 sequences from description. We removed redundant sequences above the threshold of 70% identity. We then aligned the sequences using Clustal Omega, followed by manual adjustment. We built the HMM (3.1b) using default parameter. By searching against the protein phosphatases of nine genomes, we found it matched to ~70 aa of other myotubularin subfamilies.
The region consists of PH/GRAM and phosphatase domain.
Comparison with CC1_MTM_1_PD
We aligned the alignments of CC1_MTM_1_PD and the conserved region of MTMR14. We found the conserved region of MTMR14 is a subset of CC1_MTM_1_PD.
Note: In addition to PSI-BLAST, we tried to obtain an alignment from HHPRED, but the quality is pretty bad.
