ecod_neo4j_new.txt

#Import ECOD hierarchy
CREATE CONSTRAINT ON (g:XGroup) ASSERT g.id IS UNIQUE
CREATE CONSTRAINT ON (g:HGroup) ASSERT g.id IS UNIQUE
CREATE CONSTRAINT ON (g:TGroup) ASSERT g.id IS UNIQUE
CREATE CONSTRAINT ON (g:FGroup) ASSERT g.id IS UNIQUE
CREATE CONSTRAINT ON (arch:ArchGroup) ASSERT arch.id IS UNIQUE

USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM  "file:///home/lab/Downloads/ecod118/ecod.develop118.domains.lc.txt"
AS line FIELDTERMINATOR '\t'

MATCH (dom:ECODDomain) where dom.uid = toInt(line.uid)
WITH dom, line
MERGE (arch:ArchGroup:ArchClass {id:line.arch_name}) WITH arch, dom, line

FOREACH (ignore IN CASE WHEN line.x_name <> 'no_x_name' THEN [1] ELSE [] END|
  MERGE (xgroup:XGroup:ArchClass {id: line.x_name})
  CREATE UNIQUE (arch)-[:SUBARCH]->(xgroup)
  FOREACH (ignore IN CASE WHEN line.h_name <> 'no_h_name' THEN [1] ELSE [] END|
    MERGE (hgroup:HGroup:ArchClass {id: line.h_name})
    CREATE UNIQUE (xgroup)-[:SUBARCH]->(hgroup)
    MERGE (tgroup:TGroup:ArchClass {id: line.t_name})
    CREATE UNIQUE (hgroup)-[:SUBARCH]->(tgroup)
  )
  FOREACH (ignore IN CASE WHEN line.h_name = 'no_h_name' THEN [1] ELSE [] END|
    MERGE (tgroup:TGroup:ArchClass {id: line.t_name})
    CREATE UNIQUE (xgroup)-[:SUBARCH]->(tgroup)
  )
)

FOREACH (ignore IN CASE WHEN line.x_name = 'no_x_name' THEN [1] ELSE [] END|
  FOREACH (ignore IN CASE WHEN line.h_name <> 'no_h_name' THEN [1] ELSE [] END|
    MERGE (hgroup:HGroup:ArchClass {id: line.h_name})
    CREATE UNIQUE (arch)-[:SUBARCH]->(hgroup)
    MERGE (tgroup:TGroup:ArchClass {id: line.t_name})
    CREATE UNIQUE (hgroup)-[:SUBARCH]->(tgroup)
  )
  FOREACH (ignore IN CASE WHEN line.h_name = 'no_h_name' THEN [1] ELSE [] END|
    MERGE (tgroup:TGroup:ArchClass {id: line.t_name})
    CREATE UNIQUE (arch)-[:SUBARCH]->(tgroup)
  )
)

FOREACH (f_name IN split(line.f_name, ',')|
MERGE (tgroup:TGroup {id: line.t_name})
MERGE (fgroup:FGroup:ArchClass {id: f_name})
CREATE UNIQUE (tgroup)-[:SUBARCH]->(fgroup)
CREATE UNIQUE (dom)-[:BELONGS]->(fgroup));

#Not working - here for reference
CASE WHEN line.x_name = 'no_x_name' THEN
  CASE WHEN line.h_name = 'no_h_name' THEN
    MERGE (tgroup:TGroup:ArchClass {id: line.t_name})
    CREATE UNIQUE (arch)-[:SUBARCH]->(tgroup)
  ELSE
    MERGE (hgroup:HGroup:ArchClass {id: line.h_name})
    CREATE UNIQUE (arch)-[:SUBARCH]->(hgroup)
    MERGE (tgroup:TGroup:ArchClass {id: line.t_name})
    CREATE UNIQUE (hgroup)-[:SUBARCH]->(tgroup)
  END
ELSE
  MERGE (xgroup:XGroup:ArchClass {id: line.x_name})
  CREATE UNIQUE (arch)-[:SUBARCH]->(xgroup)
  CASE WHEN line.h_name = 'no_h_name' THEN
    MERGE (tgroup:TGroup:ArchClass {id: line.t_name})
    CREATE UNIQUE (xgroup)-[:SUBARCH]->(tgroup)
  ELSE
    MERGE (hgroup:HGroup:ArchClass {id: line.h_name})
    CREATE UNIQUE (xgroup)-[:SUBARCH]->(hgroup)
    MERGE (tgroup:TGroup:ArchClass {id: line.t_name})
    CREATE UNIQUE (hgroup)-[:SUBARCH]->(tgroup)
  END
END

#Import clustering
USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM  "file:///home/lab/Downloads/ecod118/ecod.develop118.nr100.clstr.csv"
AS line FIELDTERMINATOR ','

MATCH (d:ECODDomain {uid: toInt(line.domain)}), (rep:ECODDomain {uid: toInt(line.cluster_rep)})
WITH d, rep
CREATE UNIQUE (d)-[:CLUSTER]->(rep);

#Import Ligands

CREATE CONSTRAINT ON (lig:Ligand) ASSERT lig.id IS UNIQUE

USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM  "file:///home/lab/Downloads/ecod.latest.domains.txt" 
AS line FIELDTERMINATOR '\t'

MATCH (d:ECODDomain {uid: toInt(line.uid)})

FOREACH (l IN (CASE WHEN line.ligand = 'NO_LIGANDS_4A' THEN [] ELSE split(line.ligand, ',') END)|
MERGE (lig:Ligand {id: l})
CREATE UNIQUE (d)-[:BINDS]->(lig)
);

#Import ECOD arch names
USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM  "file:///home/lab/Downloads/ecod.latest.domains.txt" 
AS line FIELDTERMINATOR '\t'

MATCH (d:ECODDomain {uid: toInt(line.uid)})
SET d.arch = line.arch_name;

USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM  "file:///home/lab/Downloads/ecod_edges.dat" 
AS line FIELDTERMINATOR ' '

MATCH (d1:ECODDomain {id: line.domain1}),(d2:ECODDomain {id: line.domain2}) 

MERGE (d1)-[:DISTANCE {id: toInt(line.key), psim: toInt(line.psim), pid: toInt(line.pid), length: toInt(line.len), rmsd: toFloat(line.rmsd)}]-(domain2);


#Import distance matrix
CREATE INDEX ON :DISTANCE(id)

USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM  "file:///home/lab/Downloads/ecod_edges.dat" 
AS line FIELDTERMINATOR ' '

MATCH (domain1:ECODDomain {id: line.domain1}), (domain2:ECODDomain {id: line.domain2}) WHERE NOT (domain1)<-[:DISTANCE]-(domain2) WITH domain1, domain2, line
CREATE UNIQUE (domain1)-[:DISTANCE {id: line.key, psim: line.psim, pid: line.pid, length: line.len, rmsd: line.rmsd}]->(domain2);


#IMPORT ECOD domains
CREATE INDEX ON :ECODDomain(id)
CREATE CONSTRAINT ON (d:ECODDomain) ASSERT d.uid IS UNIQUE
CREATE CONSTRAINT ON (g:XGroup) ASSERT g.id IS UNIQUE
CREATE CONSTRAINT ON (g:HGroup) ASSERT g.id IS UNIQUE
CREATE CONSTRAINT ON (g:TGroup) ASSERT g.id IS UNIQUE
CREATE CONSTRAINT ON (g:FGroup) ASSERT g.id IS UNIQUE
CREATE CONSTRAINT ON (p:PDBEntry) ASSERT p.id IS UNIQUE
CREATE CONSTRAINT ON (lig:Ligand) ASSERT lig.id IS UNIQUE
CREATE CONSTRAINT ON (pdbchain:PDBChain) ASSERT pdbchain.id IS UNIQUE
CREATE CONSTRAINT ON (arch:ArchGroup) ASSERT arch.id IS UNIQUE
CREATE CONSTRAINT ON (u:UniprotEntry) ASSERT u.id IS UNIQUE

USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM  "file:///home/lab/Downloads/ecod118/ecod.develop118.domains.lc.txt"
AS line FIELDTERMINATOR '\t'
WITH 
[
 pdbchain IN (split(line.seqid_range, ','))|
  [
   line.pdb+"."+toString(split(pdbchain, ':')[0]),
   [res IN split(split(pdbchain, ':')[1], '-')| res]
  ]
 ]
AS PDBCHAINRES, line

MERGE (d:ECODDomain {uid: toInt(line.uid), id: line.ecod_domain_id})

FOREACH ( sub IN PDBCHAINRES|
MERGE (pdbchain:PDBChain {id: sub[0]})
CREATE UNIQUE (d)-[:MAPPED {seqidStart: toInt(sub[1][0]), seqidEnd: toInt(sub[1][1])}]->(pdbchain)
);

#Map domains to pdbchains and uniprot
USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM  "file:///home/lab/Downloads/ecod.develop84a.uniprot_map.txt" 
AS line FIELDTERMINATOR '\t'
WITH 
[
 pdbchain IN (split(line.pdb_range, ','))|
  [
   substring(line.ecod_domain_id,1,4)+"."+toString(split(pdbchain, ':')[0]),
    CASE left(split(pdbchain, ':')[1])
      WHEN '-' THEN
        res_str = substring(split(pdbchain, ':')[1],1)
        [res IN res_str.split('-') | toInt(res)] AS tmp_res
        temp_res[0] = -1 * temp_res[0]
      ELSE
        [res IN split(split(pdbchain, ':')[1], '-')|  toInt(res)]
      END
  ]
 ] AS PDBCHAINRES, 
[
 unpres IN (split(line.unp_range, ','))|
  [res IN split(unpres,'-')| toInt(res)]
] AS UNPRES,
line

UNWIND PDBCHAINRES AS chainres
MATCH (d:Domain {uid: toInt(line.uid)}), (c:PDBChain {id: chainres[0]}) 
CREATE (d)-[:MATCHES {res: chainres[1]}]->(c)

UNWIND PDBCHAINRES AS chainres
MATCH (c:PDBChain {id: chainres[0]}), (u:UniprotEntry {accession:  line.unp_acc})
UNWIND UNPRES AS unpres
CREATE (c)-[:MATCHES {res: unpres}]->(u);

MATCH (d:Domain {uid: toInt(line.uid)}),(c:PDBChain {id: PDBCHAINRES[0]})
SET r.res = PDBCHAINRES[1]
MERGE (u:UniprotEntry {accession: line.unp_acc})
CREATE (c)-[:MATCHES {res: split(line.unp_range,'-')}]->(u);





#FIX residue mapping
MATCH (d:Domain)-[r:MATCHES]-(c:PDBChain) WHERE 3=LENGTH(r.res) SET r.res=[-1*toint(r.res[1]),toint(r.res[2])] RETURN r;

#Add group type label to groups
USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM 'file:///home/lab/Downloads/ecod.ready.latest.domains_part1.txt'
AS line FIELDTERMINATOR '\t'
WITH split(line.Fgroup, ',') AS FGROUPS, line

FOREACH (fgroup IN FGROUPS|
MATCH (f:Group {type: fgroup})
SET f:FGroup
)

MATCH (x:Group {type: line.Xgroup})
MATCH (h:Group {type: line.Hgroup})
MATCH (t:Group {type: line.Tgroup})

SET x:XGroup
SET h:HGroup
SET t:TGroup;

#FIX ARCH Hierachy
MATCH (f:FGroup)-[b1:BELONGS]-(d:Domain)-[b2:BELONGS]-(t:TGroup) CREATE UNIQUE (f)-[:BELONGS {type: "TGroup"}]->(t) DELETE b2;


#Import pdb chain to uniprot mapping from pdbsws
USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM 'file:///home/lab/Downloads/pdb_uniprot_chain_map.lst.2'
AS line FIELDTERMINATOR ','
WITH line
MATCH (c:PDBChain {id: line.pdb+"."+line.chain})

MERGE (u:Uniprot {accession: line.uniprot})
CREATE (c)-[:MAP]->(u);

#Import pdbchain to uniprot mapping from ebi SIFTS
USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM 'file:///home/lab/Downloads/pdb_chain_uniprot.csv'
AS line FIELDTERMINATOR ','
WITH line
MATCH (c:PDBChain {id: line.PDB+"."+line.CHAIN})

MERGE (u:Uniprot {accession: line.SP_PRIMARY})
CREATE (c)-[:MAP]->(u);


//RETURN [pdbchain IN (split(line.pdbRES, ','))|[subchain IN (split(pdbchain, ':'))|subchain]] AS result

//WITH split(line.pdbRES, ',') AS PDBRES
//WITH [pdbchain IN PDBRES| split(pdbchain, ':')] AS temp
//RETURN [tmp IN temp| split(tmp[1], '-')] AS result

//MERGE (d:Domain {uid: toInt(line.uid), domainID: line.domainID, status: line.status, chainID: line.chainID, pdbRES: PDBRES, assemblyStatus: line.assemblyStatus})

USING PERIODIC COMMIT 100 LOAD CSV FROM '/home/lab/Downloads/ecod.ready.latest.domains.txt' AS line FIELDTERMINATOR '\t' CREATE (d:domain {uid: toInt(line.uid), domainID: line.domainID, status: line.status, hierachyID: line.hierachyID, pdbID: line.pdbID, chainID: line.chainID, assemblyStatus: line.assemblyStatus});



LOAD CSV FROM '/home/lab/Downloads/ecod.ready.latest.domains.txt' AS line FIELDTERMINATOR '\t'
  AS line
WITH line LIMIT 4
RETURN line

#Get Single Path to Top Level Arch.
MATCH (d:Domain)-[:BINDS]-(l:Ligand {type: "CA"}),(a:ArchGroup) 
WITH d, COLLECT(a) as archs 
UNWIND archs AS arch 
MATCH p=shortestPath((d)-[:BELONGS*]-(arch)) WITH p ORDER BY length(p) 
RETURN COLLECT(p)[0] AS path;

#Find Parent
MATCH (d:Domain)-[:BINDS]-(l:Ligand {type: "CA"}),(d)-[:BELONGS*]->(a) WHERE not(a-[:BELONGS]->()) RETURN DISTINCT a;














USING PERIODIC COMMIT 100
LOAD CSV WITH HEADERS FROM  "file:///home/lab/Downloads/pdb_chain_uniprot.tsv"
AS line  FIELDTERMINATOR '\t'
WITH line LIMIT 15

MATCH (c:PDBChain {id: line.PDB+"."+line.CHAIN})
MERGE (sp:SWISSProt {id: line.SP_PRIMARY}) 
CREATE UNIQUE (c)-[m:MAPPED {seqidStart: toInt(line.RES_BEG), seqidEnd: toInt(line.RES_END), spStart: toInt(line.SP_BEG), spEnd: toInt(line.SP_END)}]->(sp)
RETURN c,m,sp;


LOAD CSV WITH HEADERS FROM  "file:///home/lab/Downloads/ecod.develop84a.uniprot_map.txt" 
AS line FIELDTERMINATOR '\t'
WITH 
[
 pdbchain IN (split(line.pdb_range, ','))|
  [
   substring(line.ecod_domain_id,1,4)+"."+toString(split(pdbchain, ':')[0]),
    CASE left(split(pdbchain, ':')[1],1)
      WHEN '-' THEN
        extract(res IN split(substring(split(pdbchain, ':')[1],1), '-')| toInt(res)) AS temp_res
        temp_res[0] = -1*temp_res[0]
      ELSE
        [res IN split(split(pdbchain, ':')[1], '-')|  toInt(res)]
      END
  ]
 ] AS PDBCHAINRES, 
[
 unpres IN (split(line.unp_range, ','))|
  [res IN split(unpres,'-')| toInt(res)]
] AS UNPRES,
line LIMIT 25

UNWIND PDBCHAINRES AS chainres
RETURN chainres[0], chainres[1];