The$Intellectual$Property$Management$ for$DataSharing$in ... · The$Intellectual$Property$Management$ for$DataSharing$in$aGerman$$ Liver$Cancer$Research$Network$ Shan$He,$Mahias$

The  Intellectual  Property  Management    for  Data  Sharing  in  a  German    Liver  Cancer  Research  Network  

Shan  He,  MaChias  Ganzinger,  Petra  Knaup  InsFtute  for  Medical  Biometry  and  InformaFcs  

Medical  InformaFcs  SecFon  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

•  efficient  and  reliable    data  storage  

•  plaLorm  for  data  sharing  •  benefits  

–  efficient  use  of  research  resources  –  avoiding  duplicate  experiments  or  trials  –  gathering  adequate  sample  sizes  –  rapid  release  of  prepublicaFon  data  

introducFon  background  

methods  results  

discussion  

IntroducFon  biomedical  research  network  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

Projektstandort:

SFB/TRR77  -­‐  Liver  Cancer  From  Molecular  Pathogenesis  to  Targeted  Therapies  

Research  Area  A  From  Chronic  Liver  Disease  to  Tumor  

FormaFon    

Research  Area  C  Novel  Approaches  to  

Therapy    

Project  Area  Z  Central  Projects  of  the  SFB/TRR77    

Project  B1  

Project  B2  

Project  B5  

Project  B3  

Project  B7  

Project  B6  

Project  B4  

Project  B8  

Project  A1  

Project  A2  

Project  A5  

Project  A3  

Project  A7  

Project  A6  

Project  A4  

Project  C1  

Project  C2  

Project  C5  

Project  C3  

Project  C7  

Project  C4  

Project  Z1  

Project  Z2  

Research  Area  B  Basic  Molecular  

Mechanisms  /  Novel  Targets    

Associated  Project  

TRR  77  is  funded    by  the  German    

Research    FoundaFon  (DFG)  

Heidelberg  Hannover  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

www.livercancer.de  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

Methods  

•  survey  among  all  22  projects  of  network  (TRR)  –  potenFal  of  data  sharing  –  concerns  

•  interviews  with  project  leaders  •  literature  review  on  promoFng  data  sharing  •  development  of  a  concept  within  a    Service  Oriented  Architecture  (SOA)  

introducFon  background  methods  

results  discussion  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

Results  

•  surveys  and  interviews  –  current  data  sharing  paCerns  

• known  colleagues  • negoFated  on  an  individual  basis  • based  on  mutual  trust  and  interest  

–  concerns  • data  confidenFality  • access  control  issues  

introducFon  background  

methods  results  

discussion  

à  difficulty  of  geang  proper  acknowledgement    for  data  contribuFon  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

no answer

21%

Who would be interested in your data?

a  number  of    TRR  projects  

57%  

all  TRR    projects  22%  

Results  Survey  (n=19,  14  projects)  

0

2

4

6

8

10

12

Would  you  provide  your  data  in  a  common  TRR  informaGon  system?  

56%  44%  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

yes,  no  sharing    11%  

yes,  access  for  all  TRR  5%  

Yes,  TRR  on  request  67%  

no,  anybody  can  access  my  

data.  6%  

no  answer  /  depends  11%  

Are  you  concerned  about  the  confidenGality  of  your  data?  

8

Results  Survey  (n=19,  14  projects)  

introducFon  background  

methods  results  

discussion  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

Literature  Review  •  intellectual  property  issues  of  data  sharing    

ànot  unique  in  biomedical  domain  

•  previous  efforts  in  biomedicine  mostly    policy  level  strategiesà  prerequisite  for  funding  

• mandatory  data  sharing  does  not  address    major  concerns  à  give  credit  to  invesFgators  

•  Idea    in  literature:  share  metadata  only    ànot  distribute  the  data  set  itself  

introducFon  background  

methods  results  

discussion  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

Key  System  Requirements  for  Intellectual  Property  Management  

• enable  data  producers  to  track  who  is  accessing  their  data  for  what  purpose  

metadata  collecFon  

• providing  data  consumers  a  formal  and  easy  way  to  acknowledge  data  producers  

data  citaFon  

• allowing  impact  of  data  set  to  be  tracked  

• establishing  policies  that  recognize  and  reward  data  producers  

incenFves  for  data  sharing  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

Metadata  CollecFon  

• metadata  for  –  data  queries  from  data  consumers  

• project  descripFon  •  research  quesFon  or  hypothesis  • expected  Fmeline  

–  data  sets  generated  by  data  producers  • brief  descripFon  of  data  source  •  related  data  sharing  policies    à  e.g.  ‚Are  review  process  and  approval  required  before  data  are  released  to  requester?‘

introducFon  background  

methods  results  

discussion  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

Data  CitaFon  

•  review  process  for  sensiFve  data  sets    à  approve  or  deny  data  requests  

•  data  consumer  has  to  agree  on    data  citaFon  statement  

•  ciFng  data  in  the  same  way    as  arFcles  and  books  –  digital  object  idenFfier  (DOI)  allocated  –  standard  metadata  scheme  for    proper  data  citaFon  

introducFon  background  

methods  results  

discussion  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

IncenFves  for  Data  Sharing  

•  tracking  the  impact  of  a  data  set    via  unique  ID  

•  establishing  policies  that  recognize  and  reward  data  producers  

•  high  impact  data  set  may  be  basis  –  funding  renewal  –  searching  for  financial  support  

introducFon  background  

methods  results  

discussion  

Medical Informatics Petra Knaup Medical Informatics Petra Knaup

workflow  for  data  registraFon  and  access  in  SFB/TRR  77  

TRR  

Medical Informatics Petra Knaup

Discussion  and  Future  Work  

•  integrate  data  citaFon  idea  from  other  domains  to  TRR  

•  novel  approach  –  easy  data  citaFon  – metadata  collecFon  –  incenFve  strategies  

•  not  sufficient  to  avoid  intenFonal  data  plagiarism  

introducFon  background  

methods  results  

discussion  

Take home message

•  data sharing in biomedical research •  researchers are concerned

about getting credit •  IT platforms for research

networks should support – metadata for

•  queries •  data sets

– easy data citation –  tracking impact of a data set