Ruth Nussinov, Ph.D.

Ruth  Nussinov, Ph.D.
Senior Investigator (Contr)
Head, Computational Structural Biology Section

Nussinov pioneered the “conformational selection and population shift” (1999) as an alternative to the “induced fit” text-book model to explain molecular mechanism of recognition and posited that population shift underlies allosteric regulation. She proposed that all conformations pre-exist and extended this pre-existing ensemble model to catalysis (2000). She also proposed that all proteins are allosteric and posited that activation by oncogenic mutations mimic the activation mechanism of the wild-type protein. She unraveled the detailed oncogenic activation of PI3K and uncovered the structural basis for oncogenic Ras signaling and its mechanistic principles. Her pioneered dynamic free energy landscape and population shift concept has contributed to extraordinary advancements in understanding structure, function and disease.

Dr. Nussinov served as the Editor-in-Chief of PLOS Computational Biology and is currently Editor-in-Chief of Current Opinion in Structural Biology.  She is an elected Fellow of the Biophysical Society, the International Society of Computational Biology (ISCB), the American Physical Society (APS) and the American Institute for Medical and Biological Engineering (AIMBE).  She is a Highly Cited Researcher (ranking among the top 3000 researchers or 1% across all fields according to Thomson Reuters Essential Science Indicators, (December 2015 and 2018) "earning them the mark of exceptional impact"). Most of her publications are conceptualized and driven by her group. She was recognized by multiple domestic and international awards, including for her work on oncogenic signaling

A personal scientific overview of her biography is at

Areas of Expertise

1) protein structure, dynamics and function, 2) signaling, 3) allosteric drugs, 4) cellular networks, 5) key oncogenic proteins, 6) computational biology

Contact Info

Ruth Nussinov, Ph.D.
Center for Cancer Research
National Cancer Institute
Building 542, Rm 603
Frederick, MD 21702-1201
Ph: 301-846-5579

Conformational Selection and Population Shift

Inspired by the milestone concept of the free energy landscape (1991), we proposed in 1999 the model of “conformational selection and population shift” as an alternative to the “induced fit” model to explain molecular recognition. According to the “induced fit” hypothesis, the initial interaction between a protein and a binding partner induces a conformational change in the protein through a stepwise process. In our “conformational selection” model it is assumed that, prior to binding, the unliganded protein exists as an ensemble of all possible conformations in dynamic equilibrium. The binding partner interacts preferentially with a weakly populated, higher-energy conformation, causing the equilibrium to shift in favor of the selected conformation. This conformation then becomes the major conformation in the complex. Importantly, there cannot be a conformational selection without a population shift. As we explained then, population shift, or the redistribution of the pre-existing conformational ensemble of the macromolecule, is the origin of allostery. A recent paper in Science noted, “Although biochemistry textbooks have championed the induced fit mechanism for more than 50 years, there is now growing support for the additional [conformational selection and population shift] binding mechanism” (quoted from Boehr DD and Wright PE.  How do proteins interact? Science 320: 1429-30, 2008).

Conformational selection has by now been observed for protein-ligand, protein-protein, protein-DNA, protein-RNA and RNA-ligand interactions. These data support the powerful new molecular recognition paradigm for processes as diverse as signaling, catalysis, gene regulation, and protein aggregation in disease. The “conformational selection and population shift” model has significantly impacted views and strategies in allosteric drug discovery, biomolecular engineering and molecular evolution. Some early references for our proposed model:  

  1. Folding funnels, binding funnels, and protein function. Tsai CJ, Kumar S, Ma B, Nussinov R. Protein Sci. 8: 1181-90, 1999.
  2. Folding funnels and binding mechanisms. Ma B, Kumar S, Tsai CJ, Nussinov R. Protein Eng. 12: 713-20, 1999.
  3. Folding and binding cascades: shifts in energy landscapes. Tsai CJ, Ma B, Nussinov R. Proc Natl Acad Sci U S A. 96:  9970-72, 1999.
  4. Folding and binding cascades: dynamic landscapes and population shifts. Kumar S, Ma B, Tsai CJ, Sinha N, Nussinov R. Protein Sci. 9: 10-9, 2000.
  5. Multiple diverse ligands binding at a single protein site: a matter of pre-existing populations. Ma B, Shatsky M, Wolfson HJ, Nussinov R. Protein Sci. 11: 184-97, 2002. 

We further proposed conformational selection and population shift in protein disorder (Structured disorder and conformational selection. Tsai CJ, Ma B, Sham YY, Kumar S, Nussinov R. Proteins  44:  418-27, 2001).

In 1999 we also proposed that the energy landscape and the “conformational selection and population shift” model are related to protein function:  “Here we extend the utility of the concept of folding funnels, relating them to biological mechanisms and function..." (Ref. 1). “Whereas previously we have successfully utilized the folding funnels concept to rationalize binding mechanisms... here we further extend the concept of folding funnels, illustrating its utility in explaining enzyme pathways, multimolecular associations, and allostery. This extension is based on the recognition that funnels are not stationary; rather, they are dynamic, depending on the physical or binding conditions.” (Ref.  4). We suggested that the energy landscape and the minima at the bottom of the funnel explain multiple binding events in signaling (Ref. 3):  “For each such [binding] event, the population around the bottom of the corresponding (folding or binding) funnel serves as the repertoire of potentially available molecules for the following binding event in the chain. As in the case of the conformers present around the bottom of the folding funnel, here, too, it is not the conformer with the highest population times that will bind in the following step. Rather, it is the conformer whose structure in the current bound stage is most favorable for the next binding event.”

In 2004 we suggested that all proteins are allosteric (Is allostery an intrinsic property of all dynamic proteins? Gunasekaran K, Ma B, Nussinov R. Proteins 57: 433-43, 2004) and, more recently, we provided the unified mechanistic underpinnings of allostery (A unified view of "how allostery works". Tsai CJ, Nussinov R. PLoS Comput Biol. 10: e1003394, 2014), the underappreciated role of allostery in the cell (The underappreciated role of allostery in the cellular network. Nussinov R, Tsai CJ, Ma B. Annu Rev Biophys. 42: 169-89, 2013), the principles of allosteric interactions in cell signaling (Principles of allosteric interactions in cell signaling. Nussinov R, Tsai CJ, Liu J. J Am Chem Soc. 136(51): 17692-701, 2014), its role in disease (Allostery in disease and in drug discovery. Nussinov R, Tsai CJ. Cell 153: 293-305, 2013), and, over the years, the design principles of allosteric drugs (e.g., Unraveling structural mechanisms of allosteric drug action. Nussinov R, Tsai CJ. Trends Pharmacol Sci. 35: 256-64, 2014; The design of covalent allosteric drugs. Nussinov R, Tsai CJ. Annu Rev Pharmacol Toxicol. 55: 249-67 2014).

Preexisting conformational ensembles in catalysis.  By 2000 we had also extended our pre-existing ensembles view to enzyme catalysis:  “The widely accepted view of enzymatic catalysis holds that there is tight binding of the substrate to the transition-state structure, lowering the activation energy. This picture may, however, be oversimplified. The real meaning of a transition state is a surface, not a single saddle point on the potential energy surface. In a reaction with a “loose” transition-state structure, the entire transition-state region, rather than a single saddle point, contributes to reaction kinetics. Consequently, here we explore the validity of such a model, namely, the enzymatic modulation of the transition-state surface. We examine its utility in explaining enzyme catalysis. We analyze the possibility that instead of optimizing binding to a well-defined transition-state structure, enzymes are optimized by evolution to bind efficiently with a transition-state ensemble, and with a broad range of activated conformations. For enzyme catalysis, the key issue is still transition state (ensemble) stabilization. The source of the catalytic power is the modulation of the transition state. However, our definition of the transition state is the entire transition-state surface rather than just a single well-defined structure. This view of the transition-state ensemble is consistent with the nature of the protein molecule, as embodied and depicted in the protein energy landscape of folding and binding funnels.” (Transition-state ensemble in enzyme catalysis: possibility, reality, or necessity? Ma B, Kumar S, Tsai CJ, Hu Z, Nussinov R. J Theor Biol. 203:  383-97, 2000). More recently, we argued that conformational transitions may involve conformational selection and induced fit, which can be viewed as a special case in the catalytic network (Enzyme dynamics point to stepwise conformational selection in catalysis. Ma B, Nussinov R. Curr Opin Chem Biol. 14: 652-9, 2010).

Protein-protein interactions.  We are very interested in protein-protein interactions. We proposed that protein folding and binding are similar processes with similar underlying mechanisms, and that the architectural motifs at protein-protein interfaces resemble those of single chain proteins (Protein-protein interfaces: architectures and interactions in protein-protein interfaces and in protein cores. Their similarities and differences. Tsai CJ, Lin SL, Wolfson HJ, Nussinov R. Crit Rev Biochem Mol Biol. 31: 127-52, 1996). This conceptual similarity, now universally accepted, led us to develop and apply similar biophysical and algorithmic approaches. In particular, it led us to propose in the 1990s and again around 2005 that interface structural similarity exists not only between homologous protein pairs; different protein fold-pairs can also interact via similar interface architectures and these architectures are similar to those observed in single chain proteins. This recognition inspired the idea that known three-dimensional (3D) interface architectures can be used as templates to identify interacting protein pairs independent of homology or global fold similarity. This has proven to be a powerful strategy, which is now followed for large-scale docking by a number of research groups.

Ras and alternative pathways in tumor proliferation.  

Recently, we took up the challenging problem of the mechanisms of Ras protein, particularly focusing on K-Ras4B-driven adenocarcinomas, its highly oncogenic isoform. We aimed to address key questions in cancer biology, including Ras regulation at the membrane, signaling through its major (MAPK and PI3K/Akt) pathways and cross-talk (e.g. via NORE1A/RASSF5 with the Hippo pathway), mechanisms of its oncogenic mutations, and more. Much of the literature had biological information as well as some structural data; we believed that viewing the problems from our conformational standpoint can help in unraveling mechanisms on the molecular basic level. Questions that we asked included - Does Ras’ disordered hypervariable region (HVR) have a role beyond membrane anchoring? Does Ras dimerize? If so, why and how? How is Ras regulated at the membrane and what are the mechanisms of Ras signaling at the membrane? What are Ras’ alternative pathways? This is highly significant, since drug resistance to Ras and its pathways in tumor proliferation can involve those pathways. Further, what are the mechanisms of Ras’ oncogenic mutations? And, of cardinal importance, calmodulin - where does it bind and how does it contribute to K-Ras4B cancer development? Further, since palmitoylation is reversible, would that alter our view of calmodulin involvement in adenocarcinomas which was attributed solely to K-Ras4B? Is trafficking from the endoplasmic reticulum to the plasma membrane via PDEδ unique to K-Ras4B or can K-Ras 4A bind as well? What is the mechanism of NORE1A (RASSF5) which links the important Hippo pathway to Ras and MAPK pathways? And most significantly, how would a newly obtained insight help in drug discovery? We aim to obtain a global cellular view of alternative pathways in tumor proliferation toward a new view of pathway-driven drug resistance in tumor proliferation. Our publications on Ras and oncogenic signaling address all of these questions, and ongoing work continues along these directions.

Ras and cancer: GTP-dependent K-Ras dimers, the role of calmodulin in KRAS-driven adenocarcinomas, the critical role of oncogenic KRAS in the initiation of cancer through deregulation of the G1 cell cycle, and a new view of Ras isoforms in cancers, the structural basis for oncogenic mutations, and more. We discovered the two major interfaces of GTP-dependent K-Ras dimers (GTP-Dependent K-Ras Dimerization. Muratcioglu S, Chavan TS, Freed BC, Jang H, Khavrutskii L, Freed RN, Dyba MA5, Stefanisko K, Tarasov SG, Gursoy A, Keskin O, Tarasova NI, Gaponenko V, Nussinov R. Structure 23(7): 1325-35, 2015). The first, highly populated β-sheet dimer interface is at the Switch I and effector binding regions, overlapping Raf’s, PI3K’s, RalGDS’ and additional effectors’ binding surfaces. This interface has to be inhibitory to such effectors. The second, helical interface also overlaps some effectors’ binding sites. This interface may promote Raf‘s activation. Our data reveal how Ras self-association can regulate effector binding and activity, and suggest that disruption of the helical dimer interface by drugs may abate Raf’s signaling in cancer.  Overall, we believe that either dimerization or nanoclustering of Ras can activate Raf.  However, Ras dimerization may assist in isoform selectivity at the membrane.  On its own, an isolated Ras monomer can also activate Raf, however we believe that the signaling output may not be at the same level.

We pointed out the overlooked critical role of calmodulin in KRAS-driven adenocarcinomas (The Key Role of Calmodulin in KRAS-Driven Adenocarcinomas). Muratcioglu S, Tsai C.-J., Jang H, Gursoy A, Keskin O, Nussinov R. Mol Cancer Res. 13(9): 1265-73, 2015). Calmodulin (CaM) selectively binds to GTP-bound K-Ras4B; but not to its isoforms. Cell proliferation and growth require the MAPK (Ras/Raf/MEK/ERK) and PI3Kα/Akt pathways. We proposed that Ca2+/CaM promote PI3K/Akt signaling, and suggest how. Ca2+/CaM involvement may explain puzzling observations like the elevated calcium levels in adenocarcinomas. We hypothesized that CaM recruits and helps activate PI3K at the membrane, and that this is the likely reason for Ca2+/CaM-dependence in adenocarcinomas. CaM can contribute to initiation/progression of ductal (pancreatic, colorectal, lung) cancers via both PI3Kα/Akt and Raf/MEK/ERK pathways. Blocking the K-Ras/MAPK pathway and CaM/PI3Kα binding in a K-Ras4B/CaM/PI3Kα trimer could be a promising adenocarcinoma-specific therapeutic strategy. While earlier data pointed to formation of a ternary complex consisting of K-Ras, P13KPI3Kα and CaM, recent data point to phosphorylated CaM binding to the SH2 domains of the p85 subunit of PI3Kα and activating it. Modeling suggests that the high affinity interaction between the phosphorylated CaM tyrosine motif and PI3Kα, can promote full PI3Kα activation by oncogenic K-Ras. We believe that that the interaction of calmodulin, phosphorylated at Tyr99, with the SH2 domains result in high affinity, and that this this is the major mode of interaction, whereas the unphosphorylated calmodulin has a lower population (the literature and proposed mechanism are reviewed at Calmodulin and PI3K Signaling in KRAS Cancers. Nussinov R, Wang G, Tsai CJ, Jang H, Lu S, Banerjee A, Zhang J, Gaponenko V. Trends Cancer. 2017 Mar;3(3):214-224. doi: 10.1016/j.trecan.2017.01.007.

Our 2015 Mol Can Res paper won an Award at the 2017 AACR meeting

Best of AACR journals 

We also proposed a new view of Ras isoforms in cancers (A New View of Ras Isoforms in Cancers. Nussinov R, Tsai C-J, Chakrabarti M, Jang H. Cancer Res. 2016 Jan 1;76(1):18-23”.  We proposed that small GTPase K-Ras4A have a single state or two states, one resembling K-Ras4B and the other N-Ras. A recent study of K-Ras4A made the remarkable observation that even in the absence of the palmitoyl K-Ras4A can be active at the plasma membrane. Importantly, this suggests that K-Ras4A may exist in two distinct signaling states. In state 1 K-Ras4A is only farnesylated, like K-Ras4B; in state 2 farnesylated and palmitoylated, like N-Ras. The K-Ras4A hypervariable region (HVR) sequence is positively charged, in-between K-Ras4B and N-Ras. Taken together, this raises the possibility that the farnesylated but nonpalmitoylated state 1, like K-Ras4B, binds calmodulin and is associated with colorectal and other adenocarcinomas like lung cancer and PDAC (pancreatic ductal adenocarcinoma). On the other hand, state 2 may be associated with melanoma and other cancers where N-Ras is a major contributor, such as acute myeloid leukemia (AML). Importantly, H-Ras has two – single and double – palmitoylated states that may also serve distinct functional roles. The multiple signaling states of palmitoylated Ras isoforms question the completeness of small GTPase Ras isoform statistics in different cancer types and call for reevaluation of concepts and protocols. They may also call for reconsideration of oncogenic Ras therapeutics.

Additional recent works on KRas include GTP Binding and Oncogenic Mutations May Attenuate Hypervariable Region (HVR)-Catalytic Domain Interactions in Small GTPase KRAS4B, Exposing the Effector Binding Site (Lu S, Banerjee A, Jang H, Zhang J, Gaponenko V, Nussinov R. J Biol Chem. 290(48): 28887-900, 2015) and Mechanisms of Membrane Binding of Small GTPase K-Ras4B Farnesylated Hypervariable Region (Jang H, Abraham SJ, Chavan TS, Hitchinson B, Khavrutskii L, Tarasova NI, Nussinov R, Gaponenko V. J Biol Chem. 2015).  We have further revealed the structural basis of oncogenic mutations in K-Ras4B (The Structural Basis of Oncogenic Mutations G12, G13 and Q61 in Small GTPase K-Ras4B. Lu S, Jang H, Nussinov R, Zhang J. Mutations in K-Ras4B are predominant at residues G12, G13 and Q61. Even though all impair GAP-assisted GTP → GDP hydrolysis, the mutation frequencies of K-Ras4B in human cancers vary. Here we aimed to figure out their mechanisms and differential oncogenicity. We performed comprehensive molecular dynamics simulations on the wild-type K-Ras4B (K-Ras4B(WT)-GTP/GDP) catalytic domain, the K-Ras4B(WT)-GTP-GAP complex, and the mutants (K-Ras4B(G12C/G12D/G12V)-GTP/GDP, K-Ras4B(G13D)-GTP/GDP, K-Ras4B(Q61H)-GTP/GDP) and their complexes with GAP. In addition, we simulated 'exchanged' nucleotide states. Our comprehensive simulations revealed that in solution K-Ras4B(WT)-GTP exists in two, active and inactive, conformations. Oncogenic mutations differentially elicit an inactive-to-active conformational transition in K-Ras4B-GTP; in K-Ras4B(G12C/G12D)-GDP they expose the bound nucleotide which facilitates the GDP-to-GTP exchange. These mechanisms may help elucidate the differential mutational statistics in K-Ras4B-driven cancers. Our simulations also revealed that the conformational transition is more accessible in the GTP-to-GDP than in the GDP-to-GTP exchange. Importantly, GAP not only donates its R789 arginine finger, but stabilizes the catalytically-competent conformation and pre-organizes catalytic residue Q61; mutations disturb the R789/Q61 organization, impairing GAP-mediated GTP hydrolysis. Together, our simulations help provide a mechanistic explanation of key mutational events in one of the most oncogenic proteins in cancer.  We have also shown that in principle PDEδ should be able to bind and translocate all Ras isoforms, not only K-Ras4B. Earlier data suggested that PDEδ extracts K-Ras4B and N-Ras from the plasma membrane (PM); but surprisingly not K-Ras4A.  PDEδ Binding to Ras Isoforms Provides a Route to Proper Membrane Localization. Muratcioglu S, Jang H, Gursoy A, Keskin O, Nussinov R. J Phys Chem B. 2017 May 25. doi: 10.1021/acs.jpcb.7b03035. Earlier analysis of the crystal structures advanced that the presence of large/charged residues adjacent to the farnesylated site precludes the PDEδ interaction. We showed that PDEδ can bind to farnesylated K-Ras4A and N-Ras like K-Ras4B - albeit not as strongly. We explained that the weaker binding to PDEδ, coupled with stronger anchoring of K-Ras4A in the membrane (but not of electrostatically neutral N-Ras), can explain the observation of why PDEδ is unable to effectively extract K-Ras4A. We thus proposed that farnesylated Ras isoforms can bind PDEδ to fulfil the required PM enrichment, and argued that the different environments - PM versus solution - can resolve apparent puzzling Ras observations. These are novel insights that would not be expected based on the crystal structures alone provide an elegant rationale for previously puzzling observations of the differential effects of PDEδ on farnesylated Ras family proteins.

Autoinhibition in Ras effectors Raf, PI3Kα, and RASSF5: a comprehensive review underscoring the challenges in pharmacological intervention. Autoinhibition is an effective mechanism that guards proteins against spurious activation. Despite its ubiquity, the distinct organizations of the autoinhibited states and their release mechanisms differ. Signaling is most responsive to the cell environment only if a small shift in the equilibrium is required to switch the system from an inactive (occluded) to an active (exposed) state. Ras signaling follows this paradigm. This underscores the challenge in drug (inhibitor) discovery to exploit and enhance autoinhibited states. Recently, we reviewed autoinhibition and release mechanisms at the membrane focusing on three representative Ras effectors, Raf protein kinase, PI3Kα lipid kinase, and NORE1A (RASSF5) tumor suppressor, and pointed to the ramifications to drug discovery. We further touched on Ras upstream and downstream signaling, Ras activation, and the Ras superfamily in this light, altogether providing a broad outlook of the principles and complexities of autoinhibition (Autoinhibition in Ras effectors Raf, PI3Kα, and RASSF5: a comprehensive review underscoring the challenges in pharmacological intervention. Nussinov R, Zhang M, Tsai CJ, Liao TJ, Fushman D, Jang H. Biophys Rev. 2018.


Ras nanoclusters. We also asked Is Nanoclustering essential for all oncogenic KRas pathways? Can it explain why wild-type KRas can inhibit its oncogenic variant?( Nussinov RTsai CJ, Jang H. Semin Cancer Biol. On-Line January 2018 ttps:// We proposed that nanoclustering (or dimerization, or oligomerization) is needed only for the Ras-driven MAPK pathway, but not the PI3K/Akt/mTOR, and that nanoclustering can explain why wild-type KRas can inhibit its oncogenic variant.

Thus, is nanoclustering essential for all oncogenic KRas pathways? We proposed that dimerization (nanoclustering) is needed only for activation of Raf/MAPK - but not PI3K. We explained that Raf’s activation requires dimerization of the catalytic domain. Thus, Raf’s activation requires Ras spatial proximity (whether as dimers oligomers, or nanoclusters). In contract, a monomeric Ras activates PI3K.


And why does wild-type Ras inhibit its oncogenic variants? And why KRas more so than NRas?

Early on, it was observed that wild-type human HRas gene can act as an onco-suppressor. Subsequent work revealed that in RAS-mutant cancers, wild-type Ras proteins can act as cancer suppressors, notably when the mutant and the wild-type are the same isoform. Shortly thereafter, it transpired that this is particularly the case for KRAS. Wild-type KRAS inhibited colony formation and tumor development in cell lines containing an actively expressed KRAS allele. Subsequent experiments confirmed and expanded on this apparent puzzling observation. Even though it was also shown to hold for other RAS isoforms, such as NRAS, the trend was not as strong. How to explain these observations, coupled with potential pharmacological implications, fueled experiments searching for answers.


Thus, how then to explain the roles of wild-type Ras proteins with respect to their corresponding oncogenic mutants in human cancers? How can the wild-type KRAS isoform act as a tumor suppressor in mutant KRAS-driven cancers? Why are the observations particularly robust for KRAS, rather than NRAS? And why a wild-type RAS isoform not suppress a mutant isoform of a different type? Further complicating the issue were observations supporting that the wild-type can promote tumorigenesis when the mutant and the wild-type belong to different isoforms, or that neither suppresses nor promotes cancer.


In the paper referenced above, we offer a simple clustering-centric interpretation, and a way to test it. Oncogenic Ras does not require an incoming signal from an RTK; wild-type Ras does. If a nanocluster of oncogenic Ras also contains inactive wild-type Ras, the MAPK signaling output of the nanocluster will be reduced. This can explain why wild-type KRAS appears to quench, or suppress, cancer by inhibiting its oncogenic form. Why then will NRAS have a more limited suppression effect than KRAS? Consider the simple statistics: KRas is more abundant in cancer cells than NRas. And why would mixed isoforms – where the wild-type and mutant are not the same – not show this suppression effect? This is because nanoclusters tend to be homogeneous, which is understandable when we consider that the membrane preferences of distinct isoforms differ. The HVR of KRas4B favors loosely packed anionic membranes in the liquid-ordered phase; NRas can anchor easily into any lipid microdomain. The paper above has the relevant references.

We further reviewed factors influencing Ras lateral diffusion, asking whether oncogenic Ras diffusion speed in the membrane is important for signaling response times and whether it affects ubiquitously all pathways. We suggested that if Ras expression is sufficiently high to dimerize (or form nanoclusters), signaling response of those pathways where dimers (or nanoclusters) are involved corresponds to the speed with which Ras molecules travel in the membrane. On average, the faster the rate at which Ras travels to dimerize, the shorter the time to MAPK signaling; but not PI3Kα. However, we argued that KRas speed may not play an important functional role because changes in mobility at this scale are unlikely to be significant. In line with this, despite the anchors' variability, lateral diffusion speeds of KRas and HRas are similar, as is that of Lck kinase; however, even though with similar anchor, Cdc42 mobility presents a different pattern, commensurate with its role in the positioning of the apical domain, suggesting that mobility evolved for function.  Nussinov RTsai CJ, Jang H Oncogenic KRas mobility in the membrane and signaling response. Semin Cancer Biol. 2018 Feb 27. pii: S1044-579X(17)30287-0. doi: 10.1016/j.semcancer.2018.02.009.

Predicting host-pathogen interactions on the structural level We further proposed that (2016) Pathogen mimicry of host protein-protein interfaces modulates immunity and have developed a structural, interface-based method to predict host-pathogen interactions (Prediction of Host-Pathogen Interactions for Helicobacter pylori by Interface Mimicry and Implications to Gastric Cancer. Guven-Maiorov E, Tsai CJ, Ma B, Nussinov R. J Mol Biol. 2017 Dec 8;429(24):3925-3941. We applied it to all human oncoviruses and constructed a network of the ‘superorganism’ interactions.

Methods. Our research group includes a strong method development component. We have developed extremely efficient algorithms with unique capabilities for structural comparisons, motif detection, and docking. These algorithms are used for prediction of binding sites on protein surfaces, detection of residue hot spots, protein classification, and protein-protein and protein-small ligand docking. Development of new algorithms is ongoing.  Mapping the Conformation Space of Wildtype and Mutant H-Ras with a Memetic, Cellular, and Multiscale Evolutionary Algorithm. Clausen R, Ma B, Nussinov R, Shehu A. PLoS Comput Biol. 2015 Sep 1;11(9):e1004470. doi: 10.1371/journal.pcbi.1004470.

NIH Scientific Focus Areas:
Cancer Biology, Computational Biology, Molecular Pharmacology, Structural Biology, Systems Biology
  1. Nussinov R, Jang H, Zhang M, Tsai CG, Sablina AA.
    Trends Cancer. May;6(5) : 369-379, 2020. [ Journal Article ]
  2. Nussinov R, Tsai CJ, Jang H
    Trends Biochem Sci. Apr 25: S0968-0004(20)30090-6, 2020. [ Journal Article ]
  3. Guven-Maiorov E, Hakouz A, Valjevac S, Keskin O, Tsai CJ, Gursoy A, Nussinov R
    J Mol Biol. May 15;432(11): 3395-3403, 2020. [ Journal Article ]
  4. Nussinov R, Jang H, Gursoy A, Keskin O, Gaponenko V
    Cell Chem. Biol. . Feb 28(2): 121-133, doi: 10.1016/j.chembiol.2020.12.012, 2021. [ Journal Article ]
  5. Zhang M, Jang H, Nussinov R.
    Chem Sci. Feb 20; 10(12): 3671-3680, 2019. [ Journal Article ]

Dr. Ruth Nussinov is a senior investigator in the Laboratory of Cancer Immunometabolism, CCR, and a professor emeritus in the Department of Human Genetics, School of Medicine, Tel Aviv University, Tel Aviv, Israel. She received her B.Sc. in microbiology from the University of Washington (Seattle, WA) and her Ph.D. in biochemistry from Rutgers University (New Brunswick, NJ). Her postdoctoral training included a fellowship at the Weizmann Institute, and two visiting scientist positions, one in the Chemistry Department at the University of California, Berkeley, and the other in the Biochemistry Department at Harvard University. Dr. Nussinov then joined the Medical School at Tel Aviv University in 1985 as an associate professor and in 1990 became a full professor. Her association with the NIH started in 1983, first with the National Institute of Child Health and Human Development and, since 1985, with the National Cancer Institute. At Tel Aviv University Dr. Nussinov directed a large group of graduate students at the Medical and Computer Science Schools in parallel to leading her research team at the NCI-CCR.  

Dr. Nussinov's 1978 paper proposed the dynamic programming algorithm for RNA secondary structure prediction. To date, this algorithm is still the leading method and is taught in bioinformatics classes in universities across Europe and in the U.S. The algorithm and its extension are described in SIAM J Appl Math, 35(1): 68-82, 1978 and Proc Natl Acad Sci U S A 77: 6309-13, 1980 (see also Biological Sequence Analysis: Probabilistic Models of Proteins and Nucleic Acids, Durbin R, Eddy SR, Krogh A, Mitchison G, eds., Cambridge University Press, 1998). Below are links to course lecture/reference materials.

Dr. Nussinov was also a pioneer in DNA sequence analysis (see Time Warps, String Edits, and Macromolecules:  The Theory and Practice of Sequence Comparison, by David Sankoff  and Joseph Kruskal, CSLI Press, 1999). In the early 1980s, she authored papers on the recurrence of nucleotide patterns, later becoming a trend in computational biology.

In 1990 Dr. Nussinov began to focus on proteins. In 1999, her NCI group proposed the model of 'conformational selection and population shift' as an alternative to ‘induced fit’ to explain molecular recognition. Biochemistry textbooks have championed the ‘induced fit’ mechanism for more than 50 years. The concept of conformational selection and population shift that her group introduced emphasized that all conformational states preexist, and that evolution has exploited them for function. This paradigm has impacted the scientific community's views and strategies in drug design, biomolecular engineering, and molecular evolution. Population shift is now broadly recognized as the origin of allostery, and thus signaling, across multimolecular complexes, pathways, and the entire cellular network. It also explains the effects of allosteric, disease-related mutations. The new concepts that we have contributed “have changed the way biophysicists and structural biologists think about protein folding, protein-protein interactions, and ligand binding” and is now included in a semester chemistry/biochemistry course in U.S. college so the students are “exposed to the depth and breadth of your work, which includes applications relevant to health such as cancer and inflammation, would be tremendously beneficial and inspiring to them”.

Her research team conducts studies of protein folding, binding and function. In particular, they investigate cancer-related proteins and pathways, aiming to figure out the mechanisms of signaling in cancer and inflammation pathways and oncogenic mutations. Dr. Nussinov’s research is interdisciplinary.

Dr. Nussinov is the author of more than 500 scientific papers and serves as editor-in-chief of PLoS Computational Biology and as an editor for the Journal of Biological Chemistry, Physical Biology, Proteins, BMC Bioinformatics, and other journals. Previously, she served two terms as editor for the Biophysical Journal.  Dr. Nussinov is also a long-term member of the NIH Macromolecular Structure and Function D (MFSD) Study Section.  She is a frequent speaker in colloquia and at international and domestic conferences and has given numerous invited talks at academic institutions around the world. She serves on site visit and grant review teams, and plays numerous roles in the scientific community.

Awards and Honors

In 2011, Dr. Nussinov was the recipient of the Biophysical Society Fellow Award “for her extraordinary contributions to advances in computational biology on both nucleic acids and proteins.” She was also elected a Fellow of the International Society of Computational Biology (ISCB) in 2013.

Course Lecture/Reference Materials

Wikipedia:  Nucleic Acid Structure

Wikiomics:  RNA Secondary Structure Prediction

RNA Structure Determination (Center for Integrative Bioinformatics, Vrije Universiteit Amsterdam)



Name Position
Hyunbum Jang Ph.D. Senior Computational Scientist (Contr.)
Tsung-Jen Liao Predoctoral Visiting Fellow (Graduate Student)
Hansaim Lim Ph.D. Postdoctoral Fellow (Visiting) (Contr.)
Ryan Maloney Ph.D. Postdoctoral Fellow (CRTA)
Emine Ozdemir Guest Researcher (Contr.)
Amarda Shehu Ph.D. Guest Researcher (Contr.)
Chung-Jung Tsai Ph.D. Programmer/Analyst IV (Contr.)
Mingzhen Zhang Ph.D. Computational Scientist (Contr.)