| SH3 |
| ID: | 1518203456 |
Query Sequence |
Download Alignment We filter this alignment to remove positions that have > 75% gaps before running GREMLIN. |
| Length: | 68 (60) | ||
| Sequences: | 7942 (4551.2) | ||
| Seq/Len: | 132.367 | ||
| Nf(neff/√len): | 587.6 |
HHblits Results: (2016_02)
Sequences per length for each position, BEFORE the coverage/gap filter was applied
Legend: Dots above the red line, indicates positions where there is > 5 sequences per length (if visible, green bar = 20 seq/len). Point of this graphic is to give you an idea if resubmitting a sub-portion of your sequence might help increase the overall Seq/Len (which after applying the coverage filter is at: 132.367).
Sequences per length for each position, AFTER the coverage/gap filter was applied
Residue pairs sorted by coevolution strength:
Legend: The darker and larger the blue dots, the higher strength in covariance. Below we provide the list of the top [1.5 x length] covarying residues with sequence seperation > 3. The i and j are positions as given in the input sequence.
| i | j | Scaled score | Prob |
| 13_A | 20_T | 3.326 | 1.00 |
| 39_L | 48_K | 2.797 | 1.00 |
| 40_G | 51_F | 2.785 | 1.00 |
| 7_A | 21_F | 2.784 | 1.00 |
| 10_D | 22_K | 2.495 | 1.00 |
| 28_I | 41_E | 2.407 | 1.00 |
| 31_E | 39_L | 2.392 | 1.00 |
| 41_E | 48_K | 2.299 | 1.00 |
| 51_F | 56_V | 2.233 | 1.00 |
| 3_P | 32_F | 2.153 | 1.00 |
| 25_D | 42_L | 2.110 | 1.00 |
| 4_W | 26_K | 1.974 | 1.00 |
| 6_T | 57_S | 1.729 | 1.00 |
| 4_W | 28_I | 1.725 | 1.00 |
| 21_F | 27_I | 1.515 | 1.00 |
| 38_W | 53_S | 1.420 | 1.00 |
| 42_L | 45_D | 1.387 | 0.99 |
| 10_D | 23_E | 1.386 | 0.99 |
| 6_T | 26_K | 1.376 | 0.99 |
| 5_A | 58_L | 1.373 | 0.99 |
| 22_K | 25_D | 1.355 | 0.99 |
| 8_L | 57_S | 1.355 | 0.99 |
| 27_I | 40_G | 1.351 | 0.99 |
| 42_L | 47_S | 1.332 | 0.99 |
| 5_A | 29_N | 1.309 | 0.99 |
| 49_G | 52_P | 1.259 | 0.99 |
| 3_P | 58_L | 1.185 | 0.98 |
| 26_K | 59_G | 1.131 | 0.97 |
| 52_P | 55_Y | 1.086 | 0.96 |
| 5_A | 56_V | 1.080 | 0.95 |
| 19_L | 40_G | 1.076 | 0.95 |
| 12_D | 20_T | 1.060 | 0.95 |
| 30_I | 41_E | 1.053 | 0.95 |
| 32_F | 38_W | 1.052 | 0.95 |
Prob = P(contact | scaled_score, seq/len).
Scaled Score = (raw_score/average(raw_scores)).
- Text file of contact predictions (all 3L/2 predictions)
- Generate restraints for ROSETTA Structure Modeling
- See coupling matrix (for protein design or mutagenesis)
- Import to GREMLIN BETA - features includes ability to upload PDB and latest P(contact) calculation based on eLife 2015 fitting to PDB data.
GREMLIN results overlayed on top 10 PDB hits:
Legend: Blue filled circles are GREMLIN results (Scaled Score > 1). The grey/red filled circles underneath are pdb residue contacts (min distance < 5 Angstroms). The shade of the circles is based on 10 HHsearch results which uses the overall probability and per-position alignment confidence. Inter oligomeric contacts in the pdb are in shades of red.
| PDB | M | Cov | P(%) | HHΔ | |
| 4n78F | 1 | 0 | 99.2 | -0.069 | Contact Map |
| 1fmkA | 1 | 1 | 99.2 | -0.036 | Contact Map |
| 2h8hA | 1 | 0.9559 | 99.1 | -0.028 | Contact Map |
| 2dvjA | 1 | 0.9853 | 99.1 | -0.021 | Contact Map |
| 1mv3A | 1 | 0.8824 | 99.1 | -0.016 | Contact Map |
| 4xi2A | 1 | 0.9853 | 99.1 | -0.015 | Contact Map |
| 3pvlA | 1 | 0.9559 | 99.1 | -0.013 | Contact Map |
| 1ri9A | 1 | 0.8382 | 99.1 | -0.01 | Contact Map |
| 4mt7A | 1 | 0 | 99.1 | -0.009 | Contact Map |
| 2pz1A | 1 | 0.8824 | 99.1 | -0.008 | Contact Map |
Cov = coverage of hit
P(%) = probability
Sequence Conservation (Image Generated using WebLogo v3)