First, download the repository:
git clone https://github.com/DunbrackLab/Kincore-standalone3
Then, create a virtual environment with the necessary packages to run kincore.
Install pixi if you haven't already:
curl -fsSL https://pixi.sh/install.sh | bash
cd Kincore-standalone3
That's it. You should now be able to run kincore inside the Kincore-standalone3 directory using pixi run kincore (see below for more details).
Note: the first time you run this command, pixi will create the environment and install all the necessary packages in the Kincore-standalone3 directory.
You can also create an anconda enviroment and install the necessary packages.
conda create --name 'kincore-standalone3' python=3.8 pandas numpy biopython hmmer --channel conda-forge --channel bioconda
Activate the virtual environment
conda activate kincore-standalone3
Go to the Kincore-standalone3 folder if you haven't already:
cd Kincore-standalone3
Note: any of the following commands can also be run using python kinase_state.py instead of pixi run kincore, provided you have the necessary packages installed (such as hmmsearch) and/or virtual environment active.
Run the help command to see available options:
pixi run kincore -h
Kincore can be run on a single mmCIF or PDB file (can contain multiple models in a single file, e.g., MD trajectories or NMR structures).
pixi run kincore 1OL5.cif
pixi run kincore 1OL5.pdb
Also works on gzipped files:
pixi run kincore 1OL5.cif.gz
pixi run kincore 1OL5.pdb.gz
To run Kincore on more than one structure, give a list of structure filenames instead (full path names recommended for robustness).
pixi run kincore list.txt
Recommended: direct the output of Kincore to a text file so you can access it later. Also, attaching "&" at the end allows kincore to run in the background.
pixi run kincore list.txt > output.txt &
First, activate your virtual environment inside the Kincore-standalone3 folder
pixi shell
Or (if using Anaconda):
conda activate kincore-standalone3
Now you can run kincore using python kinase_state.py. For example:
python kinase_state.py list.txt > output.txt &
Each kinase chain occupies 6 lines, with different data listed on each line:
1OL5 0 A Active_DFGin_BLAminus_SBin_NTin_CTin Labels family CAMK hmm CAMK score 273.7 Active DFGin BLAminus Chelix-in Saltbr-in ActLoopNT-in ActLoopCT-in Spine-in APE10-dihe-na APE9-dihe-na APE8-dihe-in APE8-rot-in APE67-dihe-in APE12-dist-in APE11-dist-in APE10-dist-in APE9-dist-in HRD-in
1OL5 0 A Active_DFGin_BLAminus_SBin_NTin_CTin Residues APEp3.E302 APEp2.I301 APE.E299 APE6.D294 APE7.L293 APE8.T292 APE9.G291 APE10.C290 APE11.L289 APE12.T288 XDFG.A273 Asp.D274 Phe.F275 Gly.G276 DFG4.W277 DFG6.V279 Lys.K162 Glu.E181 Glu4.Q185 HPN7.L196 XHRD.I253 HRD.H254 Arg.R255 HRDAsp.D256 aFasp.D311
1OL5 0 A Active_DFGin_BLAminus_SBin_NTin_CTin Distances Glu4_Phe 5.58 Lys_Phe 15.10 Lys_Glu 9.45 SaltBr 2.91 DFG6_XHRD 2.93 APE9_Arg 3.68 APE3_aFasp3 6.13 APE10_DFG4 5.27 APE11_DFG4 10.96 APE12_DFG4 8.17 Spine 3.59 3.51 3.91 3.91
1OL5 0 A Active_DFGin_BLAminus_SBin_NTin_CTin Dihedrals HRD -63.28 -60.88 Arg 73.60 1.59 XDFG -139.20 -172.08 DFG 52.22 80.16 -163.54 -3.74 DFG2 -90.83 24.66 282.61 72.55 DFG3 -49.87 -44.18 APE6 -62.23 -14.30 APE7 -43.16 -62.50 APE8 -110.77 140.24 313.06 APE9 119.45 -153.19 APE10 -153.42 162.50
1OL5 0 A Active_DFGin_BLAminus_SBin_NTin_CTin Ligands ADP:1388,MG:1389,MG:1390,MG:1394 ATPlike,Allosteric,Allosteric,Allosteric
1OL5 0 A Active_DFGin_BLAminus_SBin_NTin_CTin Actloop 26 17.70 12.06 28.30
Each line starts with a row header that identifies the structure file, model number, and chain ID, followed by the conformational state. This information is repeated on each line.
1OL5 0 A Active_DFGin_BLAminus_SBin_NTin_CTin
| Header | Meaning |
|---|---|
| 1OL5 | Structure file |
| 0 | Model number |
| A | Chain ID |
The long string in the 4th column Active_DFGin_BLAminus_SBin_NTin_CTin is the overall conformational state of the kinase chain.
| Label | Meaning |
|---|---|
| Active | The activation loop and αC-helix pass all of the criteria for active kinases |
| DFGin | The DFG Phe (or equivalent residue) is flipped "in" towards the αC-helix |
| BLAminus | The backbone dihedrals of the first three residues of the XDFG motif occupy the Ramchandran regions B, L, and A, and the χ1 dihedral of the DFG Phe sidechain (or equivalent residue) is gauche-minus |
| SBin | A salt bridge is formed between the αC-helix Glu and β3-strand Lys |
| NTin | The N-terminus of the activation loop (near the DFG motif) is hydrogen bonded to the HRD loop |
| CTin | The C-terminus of the activation loop (near the APE motif) adopts the appropriate structure for binding protein substrates |
The 5th column contains another header that tells you what kind of data follows it. Residues with structural parameters that go into determining the state of the activation loop C-terminus are labeled APEi where i is the ith residue counting backwards from the end of the APE motif.
| Label | Meaning |
|---|---|
| family CAMK | 1OL5 chain A belongs to the CAMK family |
| hmm CAMK | Membership to the CAMK family was determined via the HMM filename "CAMK.hmm" located in the "HMMs" subdirectory. There are extra HMMs for unusual kinases, such as BUB, PEAK, and TP53RK that are members of the OTHER family. |
| A | Chain ID |
| score 273.7 | HMM score of the chain A sequence for CAMK.hmm |
| Active | The activation loop and αC-helix pass all of the criteria for active kinases |
| DFGin | The DFG Phe (or equivalent residue) is flipped "in" towards the αC-helix |
| BLAminus | The backbone dihedrals of the first three residues of the XDFG motif occupy the Ramachandran regions B, L, and A, and the χ1 dihedral of the DFG Phe sidechain (or equivalent residue) is gauche-minus |
| Saltbr-in | A salt bridge is formed between the αC-helix Glu and β3-strand Lys |
| ActLoopNT-in | The N-terminus of the activation loop (near the DFG motif) is hydrogen bonded to the HRD loop |
| ActLoopCT-in | The N-terminus of the activation loop (near the DFG motif) is hydrogen bonded to the HRD loop |
| Spine-in | The regulatory spine is formed. |
| APE10dihe-na | Backbone dihedral state of the 10th residue from the end of the APE motif (in the C-terminus of the activation loop). 1OL5 chain A belongs to the CAMK family, not TYR, so dihedral criteria for this residue are not applicable to determining the Active/Inactive state. |
| APE9dihe-na | Backbone dihedral state for the 9th residue from the end of the APE motif. Labeled "NA" for the same reason as APE10dihe. |
| APE8dihe-in | Backbone dihedral state for the 8th residue from the end of the APE motif is active-like, hence the label "APE8dihe-in" (in analogy to how DFG-"in" is an active-like conformation of the DFG motif). |
| APE8rot-in | Sidechain rotamer state for the 8th residue from the end of the APE motif is active-like. |
| APE67dihe-in | Backbone dihedral states for the 6th and 7th residues from the end of the APE motif are active-like. |
| APE12-dist-in | Distance between CB of the APE12 residue and CA of the DFG4 residue is active-like. |
| APE11-dist-in | Distance between CB of the APE11 residue and CA of the DFG4 residue is active-like. |
| APE10-dist-in | Distance between CB of the APE10 residue and CA of the DFG4 residue is active-like. |
| APE9-dist-in | Distance between CA of the APE9 residue and backbone carbonyl oxygen of the HRD Arg residue is active-like. |
| HRD-in | Backbone dihedral states for the HRD His and HRD Arg residues are active-like. |
List of residues used for conformational assignments. Kincore nomenclature followed by the residue type and number in the sequence (separated by ".").
| Nomenclature | Description |
|---|---|
| APE | The end of the APE motif / the "E" of the APE (or equivalent residue). |
| APE6 | The 6th residue from the end of the APE motif, counting backwards starting from the "E" of the APE. |
| APE7 | The 7th residue from the end of the APE motif. |
| APE8 | The 8th residue from the end of the APE motif. |
| APE9 | The 9th residue from the end of the APE motif. |
| APE10 | The 10th residue from the end of the APE motif. |
| APE11 | The 11th residue from the end of the APE motif. |
| APE12 | The 12th residue from the end of the APE motif. |
| XDFG | The "X" residue immediately before the DFG motif. |
| Asp | The beginning of the DFG motif / the "D" of the DFG (or equivalent residue). |
| Phe | The 2nd residue from the beginning of the DFG motif / the "F" of the DFG (or equivalent residue). |
| Gly | The 3nd residue from the beginning DFG motif / the "G" of the DFG (or equivalent residue). |
| DFG4 | The 4th residue from the beginning DFG motif, counting forwards from the "D" of the DFG. |
| DFG6 | The 6th residue from the beginning DFG motif. |
| Lys | The conserved Lys in the β3-strand (or equivalent residue). |
| Glu | The conserved Glu in the αC-helix (or equivalent residue). |
| Glu4 | The 4th residue from the Glu in the αC-helix. |
| HPN7 | The 7th residue from the beginning of the HPN motif, located in the β4-strand (part of the regulatory spine). |
| XHRD | The "X" residue immediately before the HRD motif. |
| HRD | The conserved His (or sometimes Tyr) residue at the beginning of the HRD motif. |
| Arg | The 2nd residue from the beginning of the HRD motif / the "R" of the HRD motif (or equivalent residue). |
| HRDAsp | The 3nd residue from the beginning of the HRD motif / the "D" of the HRD motif (or equivalent residue). |
| aFasp | The conserved Asp in the αF-helix. |
List of distances used for conformational assignments.
| Nomenclature | Description |
|---|---|
| Glu4_Phe | Glu4-Cα / DFG-Phe Cζ distance for DFGin/DFGout/DFGinter calculation. Glu4 is 4 residues after the salt-bridge Glu |
| Lys_Phe | Lys-Cα / DFG-Phe Cζ distance for DFGin/DFGout/DFGinter calculation. Lys is the salt-bridge Lys |
| Lys_Glu | Lys-Cβ / Glu-Cβ distance for Chelix-in/Chelix-out calculation. |
| SaltBr | Lys-Nζ / Glu OE1,OE2 distance for SaltBr-in/SaltBr-out calculation. Minimum of distance to OE1 and OE2 |
| DFG6_XHRD | DFG6-N/O / Xhrd O/N distance for ActLoopNT-in/ActLoopNT-out calculation. Minimum of two backbone-backbone hydrogen bonds distances. |
| APE9_Arg | APE9-Cα / hRd-Arg O for ActLoopCT-in/ActLoopCT-out calculation. |
| APE10_DFG4 | APE10-Cβ / DFG4-Cα for ActLoopCT-in/ActLoopCT-out calculation for nonTYR kinases |
| APE11_DFG4 | APE10-Cβ / DFG4-Cα for ActLoopCT-in/ActLoopCT-out calculation for nonTYR kinases |
| APE12_DFG4 | APE10-Cβ / DFG4-Cα for ActLoopCT-in/ActLoopCT-out calculation for nonTYR kinases |
| Spine | Spine1-Spine2-Spine3-MaxSpine distances. All between side-chain atoms of residue pairs. Spine1 = (HRD-His, DFG-Phe), Spine2 = (DFG-Phe, Glu4), Spine3 (Glu4, HPN7). MaxSpine = max(Spine1, Spine2, Spine3) to determine if spine is broken (MaxSpine>4.5 Å) |
List of dihedral angles used for conformational assignments.
| Nomenclature | Description |
|---|---|
| HRD | φ,ψ of HRD-His in A region of Ramachandran map for active kinases |
| Arg | φ,ψ of HRD-Arg in L region of Ramachandran map for active kinases |
| XDFG | φ,ψ of X of XDFG motif. B region for BLAminus state of active kinases |
| DFG | φ,ψ,χ1,χ2 of Asp of DFG motif. L region for BLAminus state of active kinases |
| DFG2 | φ,ψ,χ1,χ2 of Phe of DFG motif. A region for BLAminus state of active kinases |
| DFG3 | φ,ψ of Gly of DFG motif. A region for BLA(A)minus state of active kinases |
| APE6 | φ,ψ of APE6 residue (6th residue from end of activation loop: X in XxxAPE). APE7,APE6 == AA or BL for active kinases |
| APE7 | φ,ψ of APE7 residue (6th residue from end of activation loop: X in XxxxAPE). APE7,APE6 == AA or BL for active kinases |
| APE8 | φ,ψ,χ1 of APE8 residue (6th residue from end of activation loop: X in XxxxxAPE). B region for active kinases. χ1 in (-120°,0°). |
| APE9 | φ,ψ of APE9 residue (6th residue from end of activation loop: X in XxxxxxAPE). B region for active TYR kinases |
| APE10 | φ,ψ of APE10 residue (6th residue from end of activation loop: X in XxxxxxxAPE). B region for active TYR kinases |
List of ligands and their types (3 or 5 letter codes from PDB, residue numbers, and types
| Nomenclature | Description |
|---|---|
| ATPlike | Any ATP-like PDB ligand (ATP, ACP, ANP, ADP, AGS) |
| Type1 | Ligand occupies ATP-binding site |
| Type1.5 | Ligand occupies ATP-binding site and part of Chelix site |
| Type2 | Ligand occupies both ATP-binding site and Chelix site |
| Type3 | Ligand occupies Chelix site |
| Allosteric | Ligand is elsewhere |
Minimum, maximum, and average of B-factors of Ca atoms of activation loop. Useful for calculating min(pLDDT) of activation loop of AlphaFold/Boltz models.
Present if there is any Ser, Thr, or Tyr in activation loop in hydrogen bonding distance of HRD-Asp