Giacomo Fiorin

Current research

I work at the crossover of biology, physics, chemistry and computer science, within the Institute for Computational Molecular Science (ICMS) directed by Michael Klein.
I use detailed models of atoms and molecules, non-equilibrium sampling methods, bioinformatics techniques and the largest supercomputers, seeking clear answers to complex problems: the self-assembly of biological matter, the flow of chemicals into the human body, and the replication of viruses and bacteria.

Advanced modeling of the human skin barrier

Structure of the lipid matrix of human stratum corneum; phase coexistence at low hydration and physiological pH; MD simulations (CHARMM36 force field) to interpret small angle X-ray and neutron scattering (SAXS and SANS) data; coarse-grained MD simulations to model the self-assembly of the lipid matrix.

Working with: Christopher MacDermaid and Michael Klein (ICMS), Russell DeVane (P&G, Inc), Sam Tonddast-Navaei (U Cincinnati), Wataru Shinoda (Nagoya University Japan), Jeffery Klauda (U Maryland).

Assembly and fusion of biological membranes

Mapping the phase coexistence in lipid membranes with mixed composition and with embedded proteins; mechanism of membrane fusion in eukaryotes and viruses; development of coarse-grained models for biologically relevant molecules.

In the figure: hypothesized intermediate of virus-cell fusion (cross section of the two membranes).

Working with: Hemant Kashyap, Shuhei Kawamoto, Christopher MacDermaid and Michael Klein (ICMS), Wataru Shinoda (Nagoya University, Japan).

Mechanism of infection by the influenza virus

Mechanism of proton conduction through the M2 viral channel; mapping drug-binding sites to guide the design of new inhibitors; predicting the fitness of viral mutations.

In the figure: density of water throughout M2's pore, seen from within the viral membrane.

Working with: Hao Dong and Michael Klein (ICMS), William DeGrado (UCSF), Lawrence Pinto and Robert Lamb (Northwestern).

Collective variables to drive molecular simulations

Design of many-body variables and algorithms to follow or to bias complex phenomena.

Development of the collective variables module (colvars), a library for NAMD and LAMMPS obtained by combining a flexible C++ code to run metadynamics, and a re-implementation by Jérôme Hénin of the ABF method.
Many different algorithms are now supported in a unified flexible interface.

In the figure: tracking the configurations of a peptide dimer by using "tilt" and "spin" variables.

Working with: Jérôme Hénin (CNRS, France), Axel Kohlmeyer (ICTP, Italy), Christopher MacDermaid (ICMS).

Teaching and outreach

Biostatistics

For 3+ year students of Biology and graduate students of the College of Science and Technology (CRN: 2926 and 2928, BIOL-3312 and BIOL-5312).

Download the course syllabus in PDF

All course material (lecture slides, practice exams, exercises) is available for download through Blackboard.

Support of molecular modeling programs

For all those questions you think I might have an intelligent answer for, feel free to contact me through the NAMD and LAMMPS mailing lists.