_ _ _ Y ‘ .
- _.-.. .;ga ‘ iféifffiffj
To: Francis Collins · =_` *.22*.;; J3? » = .
. Arnie Levine _ 2;Q€;.;.:
· , ‘. ~ · Harold Varmus _ ° ·· _ J ri It gf t ·~’e ‘ · -i* {fi i ·r ~
t i Craig Velllef . lvll iii 4 L»-Li s.
From: Eric Lander · ‘ Y - ·`· " `°i ");°i I I -` l I Q
· Re: Summary of Conversations (v.1) · _ ~ ` ·
' Date: November 14, 1999 - ` . _ _
The purpose of this memois to summarize key points that have emerged during g I
_ conversations over the past few months, to provide a framework forupcorning. . I .
. discussion-to structure a productive collaborative interaction between the public .
» Human Genome- Project and Celera Genomics. ‘ . - _i
V This version (v.1) of this memo reflects my own understanding of the _ - L,
`» conversations. If I have misunderstood anything, please feel free to correct these · ‘ i
notes, If we agree on the underlying principles, it should be possible to proceed
to structuring a» collaboration. - .
There seems to be general agreement that: ' _ ·
• The public HGP and Celera Genomics each have the capability and . · `
_ commitment to sequence thehuman genome on their own.
I . IAM NOT CONVINCED OF THIS. · ·
r _ :1 • The scientific community will be better served if we can find a viable
. .· ·’ way to join forces to produce a better product in a more timely fashion.
THIS MAY ·ONLY BE IN THE SHORT RUN
• The methods being used (clone-based and whole—genome shotgun) are,
in fact, complementary. They provide much opportunity for cross-checking.
1 ‘ TRUE, BUT ONLY IF THE PRIMARY DATA IS SHARED. 0
. • A collaboration would offer the `opportunity for joint optimization of .
‘ experimental strategy and analytical methods. · `
COLLABORATION IS THE KEY WORD HERE ‘ -
• The current appearance of antagonism and excessive competition is ._
‘ unseemly, especially for a scientific endeavor of this importance. Moreover, it is `
k ‘ " not helpful for the parties. ' ‘ . . _ · . *
THIS NOT THE FAULT OF THE HGP, BUT RATHER THE
_ V ONGOHVG PR CAMPAIGN BY CELERA/PE.. AT A · .
_ ` FUNDAMENT AL DIFFERENCE IN HOW THEDATA SHOULD BE USED. . · ‘ · . · ‘
_ _ - • The public HGP is conunitted to the complete sequence of the human
" _ genome being freely available in thepublic domain, with no resxictions i
whatsoever on its use. V - · _ _ ., . _ ‘
A) • Celera Genomics is willing to make its consensus sequence of the . A i
· ‘ human genome broadly available, but wishes to prevent commercial competitors ° _ {
‘ from repackaging this information in competing databases. This is a reasonable - ii
business interest, inasmuch as Celera has made substantial private investment in
I ‘ its sequence data. ‘ ’ `
THIS IS THEIR CHOICE. THEY NEEDN'T HAVE MADE THIS
· INVESTMENT.
. RW O 0 3 8 O_ _ _ “"
Case 3 :00-cv-00705-CFD Document 137-9 Filed 03/1 1/2005 Page 2 of 3
_' • Celera Genomics is able to see data generated by the public labs and is .· ` _ · ff. ‘ -
_ · free to use it for its proprietary databases, but ascientific publication that, . .
.. - · combines substantial data from both sources should, according to accepted ‘ ‘ ·» ‘
I · · scientific practice, be a joint publication involving authors from both groups. _ .· . ; _‘ ‘ 1
~· The outline of a collaborative effort might be as follows: I ' ‘ ` -
1. joint Publication. ‘ `
_, . 'I`he two sides would agree to joint publication, which might take the form
° of two simultaneous papers: (i) a first paper reporting analysis of the public data
from the public HCP labs, and (ii) a second collaborative paper reporting joint . · _ ‘
‘ analysis of both the public-and Celera data, co—authoréd by appropriate scientists i
` t · from both sides. . . ·;
- · THIS PAPER WOULD VALIDATE THE CELERA PRODUCT WHICH °
WOULD SIGNIFICANTLY ENHANCE I'I`S VALUE . ‘
Y* · ` ‘ _ The collaborative paper would necessarily involve that co·authors from · .
the public side have sufficient access to the Celera data and any joint analysis -
. during the preparation of a paper to be able to comfortably sign their names to a
a er. ‘ ‘
· $VIEI'HOUT CI’ED ACCESS TO T1-IE DATA I DON’T SEE HOW WE
CAN CO FORWARD. WE WOULD BE IN AN UNTENABLE POSITION. I
) CERTAINLY COULDN’FEND THE PRODUCT. AND CELERA WOULD _
CLEARLY BE THE SENIOR PARTNER. _ — _ .
4 j At the same time, Celera is concerned about the security of its privately-
r ·" > generated data and is reluctant to distribute copies. _
` We will need to work out a way to provide ac0ess—for example, by
establishing a collaborative mechanism whereby designated public HCP -
scientists are able to carry out analysis of their choosing with Celerals data`
i residing on Celera’s computers and / or on a secure third-party site. Ideally, a
positive interaction would be established leading to improved analysis and a `
better scientific paper. . V l
- ‘ There is some concem about the appearance of public HCP scientists - â€
having access to Celera's data before it is rnade public. The best solution might be
to limit the period of access, perhaps to 90 days before submission of a paper. '_
._ a NO THE CONCERN CONTINUES TO BE THB LACK OF PUBLIC ACCESS TO · `
· _ THEIR DATA; . · ·
The _most realistic timeframe would be to aim for the papers toappear A ·_ · . .
‘ near thelend of 2000 (perhaps in Science), which would submission at ` ' · ’ I
».,_ October and would involve collaboration beginning in the summer. ` · · `.
In the interim, both parties would avoid engaging in attempts to "scoopâ€
the other with publications. ‘· · -
I .
2. Data Release. . · · _ A at
The public HCP lans will continue to release their data immediately. ._
_ Upon publication of the papers, the human sequence produced from the ` _ -
analysis ofthe public and Celera data will also be released on a DVD. The DVD‘s .
package and content will reflect the fact that it isa product of both efforts.
-‘ r nw 00281
Case 3 .00-cv-00705-CFD Docu ment 137-9 Filed 03/1 1/2005 Page 3 of 3
‘ ‘ CELERA WILL RAPIDLY MAKE 'THIS OBSOLETE, SELLING 'ITIEIR . _ ‘ . _.
ENHANCED PRODUCT WHICH WILL HAVE A D SLANT · — _ g ·` Y_
_ ` · The DVD wm be avajjgblé to anyone at Tigg ygughly gg-vg;-ing fhg ' _‘ · _ _ _ · · "
physical cost of production and distribution of e DVD itselfl . ‘ · · . gg
- 'Ihe DVD will carry a shriek-wrap license that protects Celera from the ‘
.~ data pn the DVD beinglu-ansferred to and incorporated in commercial databases ' _ ~
,, competing with Ce1era's business as a database supplier. · - . . ·
I WHAT ABOUT NCBI? . _
· However, the user would not be prevented from other uses, such as: _ . . ·
• downloading the entire sequence to a local computer to it; . ‘ _ *
V _ ~ • performing scientific analysa of the data and publishing relevant . f
· ‘ portions of the data in cor•nection with reports of Such analyses: . · · =
. • using the information and Bling patent on these _ `f
X · discoveries, wi out reach·thr0ugh rights attaching. · I I
WHAT ABOUT P G CELERA POSITIONS? _ . A ’
· Users would be free, of course, to generate human genome sequence in ·
their laboratories and deposit it into Genbank . · `
. 3. Public Relations ~
3 ‘ The two sides will prepare. a document and press releases describing the
i rationale- for and nature of the collaboration. ` _ -· ‘
__ Both sides will_make all efforts to support this agreement. This includes
_ - avoiding disparaging the other party, sowing discord or undermining the ' ·
“ `¤ ```‘' collaborative spirit. It will require a generosity of spirit in acknowledging the .
legitimate and important role of the both parties.
. Placing the interaction on a positive footing should lead to better science
· and will best serve the scientific community. ' `
RW -00382
United States Free Forms
Canada Free Forms
United Kingdom Free Forms
Australia Free Forms