· Introduction/Home

· Highlights/News

· Older DSM web site
· Pasteur DSM Unit
· UMR 3528 (CNRS)

- Web Servers
· NOMAD-Ref server
· MAP_traject server
· AquaSol/PDB_Hydro
· AquaSAXS server
· Job queue status

· Papers by theme
· Papers (List)
· Older papers
· Google Scholar

· Photo du Labo

· DD-NMA (beta)


We are part of the Structural Biology and Chemistry Department in the Institut Pasteur, Paris.

We use experimental techniques such as x-ray crystallography and cryo-EM to visualize at the atomic level the structures and conformational states of large macromolecules essential to life such as
-Polymerases involved in DNA replication or DNA Repair, see some selected publications here
-membrane proteins such as ligand-gated ion channels involved in cell-cell communications, see here
-proteins implicated in nucleic acids metabolism

We complement them with MD simulations, normal modes analysis and Langevin dynamics, in order to go beyond the static pictures given by these methods and to describe (on a coarse-grained level) the transition path between functional states.

Evolutionary data (sequence alignments) must also be used to uncover what has been conserved throughout evolution and it usually illuminates the structure.

Structural Molecular Biology relies on three legs, the tripod "Structure-Dynamics-Sequences" or "Architecture-Fluctuations-Evolution", and all three are necessary for understanding the functional properties of biological macromolecules.

So we develop new computational methods to predict dynamical properties of (populations of) macromolecules (see Software, Normal Modes, below), as well as their structural heterogeneity using statistical thermodynamics.
In addition, we try to better understand the electrostatic properties of macromolecules and their interactions with the solvent and free ions, in order to be able to predict their binding properties with other molecules (proteins, substrates, allosteric modulators).
When their function is unknown, or have partners still unknown, we use sequence and genomic data to infer these properties.

Most optimisation methods that we develop are based on Mean Field theory in statistical physics (see Example).

Our main goal is to design structure-inspired drugs (pharmacology) and re-design active site(s) to make them accept other substrates (synthetic biology) or simply understand how they work at the molecular level.


Please note that our web-servers have been disconnected for external use as of April 11th, 2019, for maintenance and updates. They do work, however, internally (on campus). We will go back on-line as soon as possible.
We apologize for any inconvenience this might cause.

Publications by year (2009-2019)


-New structural evidence for an "in trans" base selection mechanism by polymerase mu in Double Strand DNA Repair, with M.R. Lieber's team (USC, CA, USA), accepted in JBC.

-An optimized method to generate large libraries of modified RNAs in Methods.

-Cryo-EM structure of the polD DNA polymerase (DP1+DP2) complex, with or without DNA, in PLoS Biol.

-An updated structure-based classification of all extant DNA polymerases (Here), including polD.

-An exploration of a multidimensional representation of amino-acids to retrieve structural information from very large sequence alignments (Here). See also a recent Review in F1000_Res.


-Molecular mechanism of proton gating in GLIC in PNAS and PDF.

-Fifth DNA Polymerases Meeting in Leiden, NL (Program here).

-CECAM Meeting on Normal Modes in IHP, Paris, September (Program here).

-New and faster calculations of Normal Modes with Patrice Koehl (Ref).

-Crystal structures of a new bacterial pentaLGIC at 2.3 A in a widely open form in PNAS and PDF.

-Positive and negative modulation of pentaLGICs by General Anesthetics in (Cell Rep. and F1000)

-Review on TdT in Current Opinion in Structural Biology (on line and F1000) and PDF.

-Design of a polymerase that generates libraries of random RNA in Nucleic Acids Res.


-Simulating the transition path between two known conformationss of a macromolecule using mixed ENMs, in J. Chem. Phys. This is a follow up of our previous MAP method (see also P. Koehl in J. Chem. Phys.)

-New methods in Normal Modes from Elastic Models (with Patrice Koehl) for automatic coarse-graining (JCTC) or at dazzling speed (Front. Mol. Bios.)

-Organisation with Y.H. Sanejouand of a one day meeting in Normal Mode analysis and Conformational Transitions in Pasteur (30 May 2017)

-String method simulation of the transition pathway for GLIC, with Pr Toby W. Allen (Melbourne, Australia, corresp. author) in PNAS. See F1000

-X-ray structures of GLIC with Barbiturates, with Pr. Trevor Smart (UCL, UK) (J. Biol. Chem.) Editor's pick, Feb 3, 2017. See also here.


DNA Polymerases and DNA Repair

-X-ray structure of Archaeal polD DNA polymerase reveals a catalytic site similar to multi-subunit RNA polymerases that are found in all domains of life, by L. Sauguet, P. Raia, G. Henneke and M. Delarue (Nature Commun). See F1000.

-Structural basis for an unexpected "in trans" templated activity by TdT: implications for V(D)J recombination and DNA double-strand-breaks repair in eukaryotes, J. Loc'h, S. Rosario and M. Delarue (Structure). See F1000.

Pentameric Ligand-gated ion channels (pentaLGICs): drug binding sites in different conformational states

-X-ray structures of GLIC with Xenon, in the open and locally-closed states, with N. Colloc'h (PLoS One).
-X-ray structures of GLIC with Bromoform and Molecular Dynamics studies with M. Baaden (Structure).


-TdT structures in complex with a DNA synapsis shed new light on DNA Double-Strand-Break Repair by NHEJ (EMBO J., Mar 2015).

-Structure of a GLIC-GlyR chimera with P.-J. Corringer (PNAS, Feb 2015)

-Structural characterization of allosteric binding sites in GLIC (Acta D, March 2015) see F1000


-Structural basis for the gating mechanism in GLIC (PNAS, Jan 2014), see F1000
For questions about the actual resolution of the closed structure see here.

-A structural perspective in the pharmacology of pLGICs (BBA, May 2014)


-Structural basis for ion permeation in GLIC (EMBO Journal, Jan 2013) at 2.4 Angstrom
-Structural basis for alcohol potentiation in mutants of GLIC with R.J. Howard (Nature Comms, April 2013) see F1000

-Snapshots of Terminal deoxynucleotidyl transferase caught in action: dynamical aspects of the two-metal-ion mechanism (J. Gouge et al., J. Mol. Biol., Jul 2013)
-Structures of inhibitors of Tdt, with G. Maga (Milan) and R. di Santo (Roma), J. Med. Chem., Sep 2013.


-Structure of Archaeal DNA polymerase (polB) from P. abyssi in editing mode by J. Gouge et al. (JMB)

-A review on cys-loop receptors with P.J. Corringer in Structure

-The locally-closed form for GLIC in 2012, see F1000


-The structure of a complex of general anesthetics with GLIC in 2011, see F1000

-AquaSAXS software and web site by F. Poitevin (Here).


-1 micro-second long MD simulation of GLIC with Marc Baaden in 2010, see F1000

-High-resolution structure of the extra-cellular domain of GLIC (Here).

-New version of AquaSol software with P. Koehl (Here).


-The atomic structure of the open form of GLIC with P.J. Corringer in 2009, see F1000

-Structure-function studies of Terminal deoxynucleotidyl transferase (Here).

-The AquaSol model was extended to include solvent-solvent interactions in PRL, see F1000


We offer several applications:

Electrostatics properties of macromolecules can be calculated in an extension of Poisson-Boltzmann theory with free ions and a dipolar solvent model, as developed with P. Koehl (UC Davis) and H. Orland (CEA)

Coarse-grained dynamical properties and Normal Modes using the Elastic Network Model were developed first as a collaboration with Y.H. Sanejouand.

Generating plausible pathways between two known end-points of a structural transition is also a collaboration with H. Orland and P. Koehl.

AquaSAXS, a web-based software to calculate SAXS spectra from PDB coordinates, including the solvent density predicted by AquaSol, is available on-line. See article. The dipolar solvent model used by AqauSAXS is described here. The web server of AquaSol and a new implementation due to P. Koehl is here.

MinActionPath (MAP) web server can be used to generate the most probable trajectory between two known structural forms of the same macromolecule (Franklin et al., 2007).

Installation of NOMAD_Ref web server (see Ref. here) was done in 2007 with Erik Lindahl and allows for the application of Normal Modes to large biological Macromolecules.

Installation of PDB_Hydro web server (see Ref. here) was done in 2007 with Erik Lindahl and implements the dipolar solvent model in Poisson-Boltzmann electrostatics. This was later replaced by AquaSol.
Note that the PDB_Hydro web server offers a number of possibilities to perform several simple homology modeling applications.

Altogether, these sites provide online servers for algorithms such as normal mode calculation. model refinement, solvation, mutations, homology modeling and transition path calculation.
The primary applications concern biological macromolecules like proteins or DNA or complexes thereof.

Go to Older site for more details on the group historical activities.

NOMAD-Ref web server
Normal Mode Analysis
NOMAD-Ref web server
Normal Mode Model Refinement
PDB_Hydro web server
Mutation & Solvation: Simple Modeling
PDB_Hydro web server
AquaSaxs web server

  Marc Delarue http://lorentz.dynstr.pasteur.fr