Skip to main content

​
Overview

HHR (HH-suite Results) is the standard output format for hhblits, hhsearch, and hhalign. It contains search statistics, hit lists, and detailed pairwise alignments.

​
Format Specification

​
Header Section

Query information and search parameters: 
Query         sp|Q5VUD6|FA69B_HUMAN Protein FAM69B OS=Homo sapiens GN=FAM69B PE=2 SV=3
Match_columns 431
No_of_seqs    1 out of 1
Neff          1.0 
Searched_HMMs 1000
Date          Fri Jul 26 12:18:42 2013
Command       ./hhblits -i ../data/query.seq -d /home/meiermark/databases/uniprot20_2013_03/uniprot20_2013_03 -cpu 4 -n 1

​
Header Fields

  • Query: Query sequence identifier and description
  • Match_columns: Number of match states in query HMM
  • No_of_seqs: Number of sequences in query alignment
  • Neff: Effective number of sequences (diversity measure)
  • Searched_HMMs: Number of database HMMs searched
  • Date: Search timestamp
  • Command: Full command line used

​
Summary Hit List

Tabular list of significant hits: 
 No Hit                             Prob E-value P-value  Score    SS Cols Query HMM  Template HMM
  1 UP20|TITVEVABA|104|471 Protein 100.0  4E-171  3E-176 1228.9   0.0  413   19-431    58-470 (471)
  2 UP20|REZTOXABA|10|392 LOC10003 100.0  1E-142  1E-147 1017.2   0.0  366   66-431    27-392 (392)
  3 UP20|DUSGUVABA|1|331 Uncharact 100.0 6.2E-67 3.8E-72  487.7   0.0  314   19-332    16-329 (331)

​
Summary Columns

  1. No: Rank number
  2. Hit: Target identifier and description
  3. Prob: Probability of true positive (0-100%)
  4. E-value: Expected number of false positives with this score
  5. P-value: Probability of false positive
  6. Score: Raw alignment score
  7. SS: Secondary structure score contribution
  8. Cols: Number of aligned columns
  9. Query HMM: Query range (start-end)
  10. Template HMM: Template range (start-end, total length)

​
Detailed Alignments

For each significant hit, a detailed alignment is provided: 
No 1
>UP20|TITVEVABA|104|471 Protein FAM69B OS=Homo sapiens GN=FAM69B PE=2 SV=3
Probab=100.00  E-value=4e-171  Score=1228.89  Aligned_cols=413  Identities=100%  Similarity=2.485  Sum_probs=405.8  Template_Neff=1.000

Q sp|Q5VUD6|FA69    19 RLQGRLPGLRVRCIFLAWLGVFAGSWLVYVHYSSYSERCRGHVCQVVICDQYRKGIISGSVCQDLCELHMVEWRTCLSVAPG  98
Q Consensus        19 rlqgrlpglrvrciflawlgvfagswlvyvhyssysercrghvcqvvicdqyrkgiisgsvcqdlcelhmvewrtclsvapg  98
                      ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
T Consensus        58 RLQGRLPGLRVRCIFLAWLGVFAGSWLVYVHYSSYSERCRGHVCQVVICDQYRKGIISGSVCQDLCELHMVEWRTCLSVAPG 137
T UP20|TITVEVABA   58 RLQGRLPGLRVRCIFLAWLGVFAGSWLVYVHYSSYSERCRGHVCQVVICDQYRKGIISGSVCQDLCELHMVEWRTCLSVAPG 137

​
Alignment Header

  • Probab: Probability that hit is a true positive (0-100%)
  • E-value: Expected number of false positives
  • Score: Raw alignment score (bits)
  • Aligned_cols: Number of aligned columns
  • Identities: Percent sequence identity
  • Similarity: Average similarity score
  • Sum_probs: Sum of posterior probabilities
  • Template_Neff: Diversity of template MSA

​
Alignment Block

Four-line format:
  1. Q: Query sequence with position range
  2. Q Consensus: Query consensus sequence
  3. Match line: Shows conservation (|=identical, +=similar, .=weakly similar, space=different)
  4. T Consensus: Template consensus sequence
  5. T: Template sequence with position range
Optional additional lines:
  • Q ss_pred: Query secondary structure prediction
  • T ss_pred: Template secondary structure prediction
  • Q ss_dssp: Query DSSP secondary structure
  • T ss_dssp: Template DSSP secondary structure
  • Confidence: Alignment confidence scores

​
Example Full Output

Query         sp|Q5VUD6|FA69B_HUMAN Protein FAM69B
Match_columns 431
No_of_seqs    149 out of 270
Neff          5.2
Searched_HMMs 10000
Date          Wed Jan  4 15:14:55 2012
Command       hhblits -i query.a3m -d database -n 2

 No Hit                             Prob E-value P-value  Score    SS Cols Query HMM  Template HMM
  1 PROTEIN_A                       99.9 1.2E-35 8.5E-41  245.6   0.0  210   15-225    10-220 (450)
  2 PROTEIN_B                       95.3 0.00012 8.3E-09  125.4   0.0  180   50-230    25-205 (380)

No 1
>PROTEIN_A Full description here
Probab=99.90  E-value=1.2e-35  Score=245.60  Aligned_cols=210  Identities=45%  Similarity=1.234

Q sp|Q5VUD6|FA69    15 MRRLRRLAHLVLFCPFSKRLQGRLPGLRVRCIFLAWLGVFAGSWLVYVHYSSYSERCRGH  74
Q Consensus        15 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~gh  74
Q ss_pred             CCCCCCCHHHHHHHHHHHHHHHHCCCEEEEEEEEECCCCEEEEEEEECCCCCCCCCCCC
                      ||||||||||| |||||||||| ||||||||||||||||||||||||||||||||||
T Consensus        10 MRRLRRLAHLVIFCPFSKRLQGRLPGLRVRCIFLAWLGVFAGSWLVYVHYSSYSERCRGH  69
T PROTEIN_A        10 MRRLRRLAHLVIFCPFSKRLQGRLPGLRVRCIFLAWLGVFAGSWLVYVHYSSYSERCRGH  69
T ss_pred             CCCCCCCHHHHHHHHHHHHHHHHCCCEEEEEEEEECCCCEEEEEEEECCCCCCCCCCCC
Confidence            99999999999999999999999999999999999999999999999999999999999

​
Score Interpretation

​
Probability

  • >95%: Very confident hit, likely homologous
  • 50-95%: Possible homolog, requires validation
  • <50%: Low confidence, possibly spurious

​
E-value

  • <0.001: Highly significant
  • 0.001-0.1: Significant
  • 0.1-1: Marginally significant
  • >1: Not significant

​
Score

  • Depends on alignment length
  • Higher is better
  • Compare similar-length alignments

​
Parsing HHR Files

​
Extract Hit List

# Get top 10 hits
grep "^\s*[0-9]" results.hhr | head -10

​
Extract E-values

# Get E-values for all hits
grep "E-value=" results.hhr

​
Extract Alignments

# Get alignment for hit No 1
sed -n '/^No 1$/,/^No 2$/p' results.hhr

​
Customizing Output

​
Controlling Number of Hits

# Show top 100 hits
hhsearch -i query.a3m -d database -Z 100 -B 100

# Show minimum 20 hits
hhsearch -i query.a3m -d database -z 20 -b 20

​
E-value Threshold

# Only show hits with E-value < 0.01
hhsearch -i query.a3m -d database -E 0.01

​
Probability Threshold

# Only show hits with probability > 90%
hhsearch -i query.a3m -d database -p 90

​
Alternative Output Formats

​
BLAST Tabular Format

hhblits -i query.a3m -d database -blasttab output.m8
Columns: query, target, #match/tLen, #mismatch, #gapOpen, qstart, qend, tstart, tend, eval, score

​
A3M Alignment

hhblits -i query.a3m -d database -oa3m result.a3m
Merged alignment of significant hits in A3M format.

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us