Align A Sequence Against A Pre-Made Alignment?

Thanks in advance for any advice.

I am working on predicting structures for vhh antibody molecules and so I am exploring different options for the MSA step.

Assuming I produced a .a3m alignment file for a query sequence. Also that the query sequence is similar to a secondary sequence that I have. Would it be possible to align the secondary sequence to the previously made alignment? Primarily I want to produce an alignment file for the secondary sequence in a more efficient way since I already have an alignment for a similar sequence.

I saw that in a previous question the MUSCLE --profile flag was mentioned, but that seems to align two separate alignments rather than a single sequence to an entire alignment. Any direction for this project would be greatly appreciated!

python msa clustalw

• 1,899 views • link

updated 3.1 years ago by Mensur Dlakic

• written 3.1 years ago by BioGrad321 •

Link to previous post discussing MUSCLE Profile option. Here

• 0 views • link

written 3.1 years ago by BioGrad321 •

1 answer

You could make a hidden Markov model (HMM) from the alignment using hmmbuild and then align any number of sequences you wish to that HMM using hmmalign. It is literally typing two commands with the files you already have. You will need to download the HMMer suite and there is extensive documentation about all commands that are available.

• 0 views • link

written 3.1 years ago by Mensur Dlakic

Hey thanks for the advice!

I am trying to complete this in hhsuite since my files are stored as .a3m files which are incompatible with HMMER. It seems to have all the same functions that you mentioned but I am not sure if the functionality is the exact same. Using hhsuite I was able to produce hidden Markov model from the MSA I generated, but when I used hhsuitealign, the resulting .hhr file was empty. Do you know if it is possible to use hhsuite for this task or is HMMER required.

• 0 views • link

written 3.1 years ago by BioGrad321 •

In HHsuite the alignment program is used to align two HMMs, not an HMM and a single sequence. You can try making an HMM from your single sequence, and then it might work.

hmmbuild can make HMMs from .a3m files, but you may need to remove secondary structure predictions or any other additions beyond protein sequences.

hmmbuild -n PLMP PLMP.hmm PLMP.a3m

# hmmbuild :: profile HMM construction from multiple sequence alignments
# HMMER 3.3 (Nov 2019); http://hmmer.org/
# Copyright (C) 2019 Howard Hughes Medical Institute.
# Freely distributed under the BSD open source license.
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
# input alignment file:             PLMP.a3m
# output HMM file:                  PLMP.hmm
# name (the single) HMM:            PLMP
# - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

# idx name                  nseq  alen  mlen eff_nseq re/pos description
#---- -------------------- ----- ----- ----- -------- ------ -----------
1     PLMP                  1048   199    64    14.27  0.875

# CPU time: 0.10u 0.01s 00:00:00.11 Elapsed: 00:00:00.11

more PLMP.a3m

>PLMP_domain
VAGGNACVFVLDKAGRPLMPTRPSRARTLLRDKKAVVARYHPFTIRLKDRSNGVVQPLALLLDP
>UniRef100_A0A0E3WRW9 RRXRR domain-containing protein n=1 Tax=Methanosarcina lacustris Z-7289 TaxID=1434111 RepID=A0A0E3WRW9_9EURY
-----MLVFVINQNKKPLMPCKPSIARKLLQAGKAKVVRNTPFTIKLLFGSSGYTQPVIAGMDT
>UniRef100_A0A0F2S8G3 HNH endonuclease (Fragment) n=1 Tax=Peptococcaceae bacterium BRH_c23 TaxID=1629714 RepID=A0A0F2S8G3_9FIRM
-----MLTYVLNQNGKPLMPCKSSKARRLLKQSKAKVVKLEPFTLQLLHGSSGYKQEITLGVDA
>UniRef100_A0A0P9D4H4 RRXRR domain-containing protein (Fragment) n=1 Tax=Kouleothrix aurantiaca TaxID=186479 RepID=A0A0P9D4H4_9CHLR
-----PFIPVISANGKRLMPTTNRKADRLIASGRAlRRFSRGLFYIQLTDRTDGYTQPIAVGIDP
>UniRef100_A0A101HUK0 HNH endonuclease n=2 Tax=Petrotoga TaxID=28236 RepID=A0A101HUK0_9BACT
-----MLVYVLNKHGKPLMPCKPSKARKLLKDGKAKVVRKEPFTIQLLYGSSGYKQPITLGIDA
>UniRef100_A0A1H9BN62 RRXRR protein n=1 Tax=Ectothiorhodospira magna TaxID=867345 RepID=A0A1H9BN62_9GAMM
-----MAVLVLDKRKKPLMPCSEKRARLLLERGQAVVHRMHPFTIRLKDRTveESVLQPIQIKIDP
>UniRef100_A0A2N1DSF1 HNH endonuclease (Fragment) n=1 Tax=Psychromonas sp. MB-3u-54 TaxID=2058319 RepID=A0A2N1DSF1_9GAMM
----------------------------LLKQGKAKVKKRMPFTIKMVEDTTEFIQPIIGGMDT
>UniRef100_A0A2T3J7W4 HNH endonuclease (Fragment) n=2 Tax=Photobacterium phosphoreum TaxID=659 RepID=A0A2T3J7W4_PHOPO
-----MLVFVLNKNKEPLMPCLPSKARKLLKDGKAKVVKRTPFTITLLHGSSGYKQPLVAGMDT
>UniRef100_A0A416KUW5 HNH endonuclease (Fragment) n=1 Tax=Clostridium sp. AF27-2AA TaxID=2292206 RepID=A0A416KUW5_9CLOT
------MVYVLDKNGQPLMPTDRhRKVRLMLQSGQAKVIKRCPFTIQLNYDSGHQTQEISLGIDA
>UniRef100_A0A455V282 RRXRR domain-containing protein n=1 Tax=Halomonas axialensis TaxID=115555 RepID=A0A455V282_9GAMM
-----MAIFVLDKKKQPLMPCSEKRARLLLARGRAVVHKRYPFTIRLKDRVGGETQPLRLGIDP
>UniRef100_A0A5C9EYS1 HNH endonuclease n=1 Tax=Candidatus Lokiarchaeota archaeon TaxID=2053489 RepID=A0A5C9EYS1_9ARCH
-----MQVYVLNEDGTPLMPCVPVIARLLLKEGKARVVMRCPFTVMLTYHPaTEHVQPLTLGQDA

• 0 views • link

written 3.1 years ago by Mensur Dlakic