Automatic compilation of MATLAB programs for synergistic execution on heterogeneous processors

MATLAB is an array language, initially popular for rapid prototyping, but is now being increasingly used to develop production code for numerical and scientific applications. Typical MATLAB programs have abundant data parallelism. These programs also have control flow dominated scalar regions that have an impact on the program's execution time. Today's computer systems have tremendous computing power in the form of traditional CPU cores and throughput oriented accelerators such as graphics processing units(GPUs). Thus, an approach that maps the control flow dominated regions to the CPU and the data parallel regions to the GPU can significantly improve program performance. In this paper, we present the design and implementation of MEGHA, a compiler that automatically compiles MATLAB programs to enable synergistic execution on heterogeneous processors. Our solution is fully automated and does not require programmer input for identifying data parallel regions. We propose a set of compiler optimizations tailored for MATLAB. Our compiler identifies data parallel regions of the program and composes them into kernels. The problem of combining statements into kernels is formulated as a constrained graph clustering problem. Heuristics are presented to map identified kernels to either the CPU or GPU so that kernel execution on the CPU and the GPU happens synergistically and the amount of data transfer needed is minimized. In order to ensure required data movement for dependencies across basic blocks, we propose a data flow analysis and edge splitting strategy. Thus our compiler automatically handles composition of kernels, mapping of kernels to CPU and GPU, scheduling and insertion of required data transfer. The proposed compiler was implemented and experimental evaluation using a set of MATLAB benchmarks shows that our approach achieves a geometric mean speedup of 19.8X for data parallel benchmarks over native execution of MATLAB.

Noniterative inversion strategy for photoacoustic tomography: recovery of absorbed energy map from boundary pressure measurements

The inverse problem in photoacoustic tomography (PAT) seeks to obtain the absorbed energy map from the boundary pressure measurements for which computationally intensive iterative algorithms exist. The computational challenge is heightened when the reconstruction is done using boundary data split into its frequency spectrum to improve source localization and conditioning of the inverse problem. The key idea of this work is to modify the update equation wherein the Jacobian and the perturbation in data are summed over all wave numbers, k, and inverted only once to recover the absorbed energy map. This leads to a considerable reduction in the overall computation time. The results obtained using simulated data, demonstrates the efficiency of the proposed scheme without compromising the accuracy of reconstruction.

The Escherichia coli Phosphotyrosine Proteome Relates to Core Pathways and Virulence

While phosphotyrosine modification is an established regulatory mechanism in eukaryotes, it is less well characterized in bacteria due to low prevalence. To gain insight into the extent and biological importance of tyrosine phosphorylation in Escherichia coli, we used immunoaffinity-based phosphotyrosine peptide enrichment combined with high resolution mass spectrometry analysis to comprehensively identify tyrosine phosphorylated proteins and accurately map phosphotyrosine sites. We identified a total of 512 unique phosphotyrosine sites on 342 proteins in E. coli K12 and the human pathogen enterohemorrhagic E. coli (EHEC) O157:H7, representing the largest phosphotyrosine proteome reported to date in bacteria. This large number of tyrosine phosphorylation sites allowed us to define five phosphotyrosine site motifs. Tyrosine phosphorylated proteins belong to various functional classes such as metabolism, gene expression and virulence. We demonstrate for the first time that proteins of a type III secretion system (T3SS), required for the attaching and effacing (A/E) lesion phenotype characteristic for intestinal colonization by certain EHEC strains, are tyrosine phosphorylated by bacterial kinases. Yet, A/E lesion and metabolic phenotypes were unaffected by the mutation of the two currently known tyrosine kinases, Etk and Wzc. Substantial residual tyrosine phosphorylation present in an etk wzc double mutant strongly indicated the presence of hitherto unknown tyrosine kinases in E. coli. We assess the functional importance of tyrosine phosphorylation and demonstrate that the phosphorylated tyrosine residue of the regulator SspA positively affects expression and secretion of T3SS proteins and formation of A/E lesions. Altogether, our study reveals that tyrosine phosphorylation in bacteria is more prevalent than previously recognized, and suggests the involvement of phosphotyrosine-mediated signaling in a broad range of cellular functions and virulence.

Higher-Dimensional Analogues of the Map Coloring Problem

After a brief discussion of the history of the problem, we propose a generalization of the map coloring problem to higher dimensions.

A Multicast Authentication Protocol (MAP) for Dynamic Multicast Groups in Wireless Networks

In this paper, we have proposed a centralized multicast authentication protocol (MAP) for dynamic multicast groups in wireless networks. In our protocol, a multicast group is defined only at the time of the multicasting. The authentication server (AS) in the network generates a session key and authenticates it to each of the members of a multicast group using the computationally inexpensive least common multiple (LCM) method. In addition, a pseudo random function (PRF) is used to bind the secret keys of the network members with their identities. By doing this, the AS is relieved from storing per member secrets in its memory, making the scheme completely storage scalable. The protocol minimizes the load on the network members by shifting the computational tasks towards the AS node as far as possible. The protocol possesses a membership revocation mechanism and is protected against replay attack and brute force attack. Analytical and simulation results confirm the effectiveness of the proposed protocol.

Incorporation of Interfacial Intermetallic Morphology in Fracture Mechanism Map for Sn-Ag-Cu Solder Joints

A fracture mechanism map (FMM) is a powerful tool which correlates the fracture behavior of a material to its microstructural characteristics in an explicit and convenient way. In the FMM for solder joints, an effective thickness of the interfacial intermetallic compound (IMC) layer (t (eff)) and the solder yield strength (sigma (ys,eff)) are used as abscissa and ordinate axes, respectively, as these two predominantly affect the fracture behavior of solder joints. Earlier, a definition of t (eff), based on the uniform thickness of IMC (t (u)) and the average height of the IMC scallops (t (s)), was proposed and shown to aptly explain the fracture behavior of solder joints on Cu. This paper presents a more general definition of t (eff) that is more widely applicable to a range of metallizations, including Cu and electroless nickel immersion gold (ENIG). Using this new definition of t (eff), mode I FMM for SAC387/Cu joints has been updated and its validity was confirmed. A preliminary FMM for SAC387/Cu joints with ENIG metallization is also presented.

Insertion Reactions of Six-Membered Cyclopalladated N,N `,N `'-Triarylguanidine, Pd{kappa(2)(C,N)-C6H3Me-3(NHC(NHAr)(=NAr))-2)(mu-Br)](2) (Ar=2-MeC6H4) with PhCC C-C(O)OR (R = Me and Et): A Gateway to Second Orthopalladation through Novel Rearrangements

The reaction of Pd{kappa(2)(C,N)-C6H3Me-3-(NHC(NHAr)(=NAr))-2}(mu-Br)](2) (Ar = 2-MeC6H4; 1) with 4 equiv of PhC C-C(O)OMe in CH2Cl2 afforded Pd{kappa(2)(C,N)-C(Ph)=C(C(O)OMe)C(Ph)=C(C(O)-OMe)C6H3Me-3(N=C(NH Ar)(2))-2}Br] (Ar = 2-MeC6H4; 2) in 70% yield, and the aforementioned reaction carried out with 10 equiv of PhC C-C(O)OR (R = Me, and Et) afforded an admixture of two regioisomers of Pd{kappa(3)(N,C,O)-O=C(OR)-C5Ph3(C(O)OR)C(C(O)OR)C6H3Me-3(N=C(NHAr)( 2))- 2}Br] (Ar = 2-MeC6H4; R = Me (3a/3b), Et (4a/4b)) in 80 and 87% yields, respectively. In one attempt, the minor regioisomer, 4b, was isolated from the mixture in 6% yield by fractional crystallization. Palladacycles 3a/3b and 4a/4b, upon stirring in CH2Cl2/MeCN (1/1, v/v) mixture at ambient condition for S days, afforded Pd{eta(3)-allyl,(KN)-N-1)-C-5(C(O)OR)(2)Ph3C-(C(O)OR)C6H3Me-3(N=C(NH Ar)(2))(-2)}Br] (Ar = 2-MeC6H4; R = Me (5a/5b), Et (6a/6b)) in 94 and 93% yields, respectively. Palladacycles 3a/3b and 4a/4b, upon reaction with AgOTf in CH2CH2/Me2C(O) (1/1, v/v) mixture at ambient temperature for 15 min, afforded Pd{kappa(3)(N,C,O)-O=C(OR)C5Ph3(C(O)OR)C(C(O)OR)C6H3Me-3(N=C(NHAr)(2 ))-2}(OTf)] (Ar = 2-MeC6H4; R = Me (7a/7b), Et (8a/8b)) in 79 and 77% yields, respectively. Palladacycles 7a/7b and 8a/ 8b, upon reflux in PhC1 separately for 6 h, or palladacycles 5a/5b and 6a/6b, upon treatment with AgOTf in CH2Cl2/Me2C(O) (7/3, v/v) mixture for 15 min, afforded Pd{(eta(2)-Ph)C5Ph2(C(O)OR)kappa(2)(C,N)-C(C(O)OR)C6H3Me-3(N=C(NHAr) (2))-2}(OTf)] (Ar = 2-MeC6H4; R = Me (9a/9h), Et (10a/10b)) in >= 87% yields. Palladacycles 9a/9b, upon stirring in MeCN in the presence of excess NaOAc followed by crystallization of the reaction mixture in the same solvent, afforded Pd{kappa(3)(N,C,C)-(C6H4)C5Ph2(C(O)OMe)(2)C(C(O)OMe)(2)C6H3Me-3(N=C( NHAr)(2))-2}(NCMe)] (Ar = 2-MeC6H4; 11a/11b) in 82% yield. The new palladacycles were characterized by analytical, IR, and NMR (H-1 and C-13) spectroscopic techniques, and the molecular structures of 2, 3a, 4a, 4b, 5a, 6a, 7a, 9a, 10a, and 11a-d(3) were determined by single crystal X-ray diffraction. The frameworks in the aforementioned palladacycles, except that present in 2, are unprecedented. Plausible pathways for the formation of new palladacycles and the influence of the guanidine unit in 1, substituents in alkynes, reaction conditions, and electrophilicity of the bromide and the triflate upon the frameworks of the insertion products have been discussed.

Imaging cellular signalling: many `moving tales' in MAP kinase odyssey

Cellular signalling events are at the core of every adaptive response. Signalling events link environmental changes to physiological responses, consequently allowing cellular and organismal sustenance and survival. Classical approaches to study cellular signalling have relied on a variety of cell disruptive techniques which yield limited kinetic information, while the underlying events are much more complex. In this article, we discuss how modern live cell imaging microscopy has found increasing utilization in revealing spatio temporal dynamics of various signalling pathways. Utilizing the well studied mitogen-activated protein kinase (MAPK) signalling cascade as a template, the design, construction and utilization of `mobile' (translocation proficient) biosensors, suitable for studying MAPK signalling in living cells are described in detail. Experimental setup and results obtained from these biosensors, based on different proteins involved in the MAPK signalling cascade, have been described along with the setup of a microscope optimal for live cell imaging applications. Utilizing the ability to activate or deactivate signalling pathways using defined activators and specific pharmacological inhibitors, we also show how these sensors can yield unique spatial and temporal kinetic information of signalling in living cells.

Magnetic Structure and Properties of the Rechargeable Battery Insertion Compound Na2FePO4F

The magnetic structure and properties of sodium iron fluorophosphate Na2FePO4F (space group Pbcn), a cathode material for rechargeable batteries, were studied using magnetometry and neutron powder diffraction. The material, which can be described as a quasi-layered structure with zigzag Fe-octahedral chains, develops a long-range antiferromagnetic order below similar to 3.4 K. The magnetic structure is rationalized as a super-exchange-driven ferromagnetic ordering of chains running along the a-axis, coupled antiferromagnetically by super-super-exchange via phosphate groups along the c-axis, with ordering along the b-axis likely due to the contribution of dipole dipole interactions.

Representative seismic hazard map of Coimbatore, India

The seismic hazard value of any region depends upon three important components such as probable earthquake location, maximum earthquake magnitude and the attenuation equation. This paper presents a representative way of estimating these three important components considering region specific seismotectonic features. Rupture Based Seismic Hazard Analysis (RBSHA) given by Anbazhagan et al. (2011) is used to determine the probable future earthquake locations. This approach is verified on the earthquake data of Bhuj region. The probable earthquake location for this region is identified considering earthquake data till the year 2000. These identified locations match well with the reported locations after 2000. The further Coimbatore City is selected as the study area to develop a representative seismic hazard map using RBSHA approach and to compare with deterministic seismic hazard analysis. Probable future earthquake zones for Coimbatore are located considering the rupture phenomenon as per energy release theory discussed by Anbazhagan et at (2011). Rupture character of the region has been established by estimating the subsurface rupture length of each source and normalized with respect to the length of the source. Average rupture length of the source with respect to its total length is found to be similar for most of the sources in the region, which is called as the rupture character of the region. Maximum magnitudes of probable zones are estimated considering seismic sources close by and regional rupture character established. Representative GMPEs for the study area have been selected by carrying out efficacy test through an average log likelihood value (LLH) as ranking estimator and considering the Isoseismal map. New seismic hazard map of Coimbatore has been developed using the above regional representative parameters of probable earthquake locations, maximum earthquake magnitude and best suitable GMPEs. The new hazard map gives acceleration values at bedrock for maximum possible earthquakes. These results are compared with deterministic seismic hazard map and recently published probabilistic seismic hazard values. (C) 2014 Elsevier B.V. All rights reserved.

NON-GLYCINE RESIDUES IN THE DISALLOWED REGIONS OF THE RAMACHANDRAN MAP AND THEIR NEIGHBORHOOD PREFERENCES

The allowed and the ``disallowed'' regions in the celebrated Ramachandran map (phi-psi] map) was elegantly deduced by Ramachandran, Ramakrishnan and Sasisekharan even before the protein crystal structures became available. This powerful map was derived based on rigid geometry of the peptide group and later several investigations on protein crystal structures reported the occurrence of a small fraction of the phi-psi] torsion angles in the disallowed region. The question is what factors make these residues adopt disallowed conformations? Is it driven by the necessity to maintain the overall topology or is it associated with function or is it just that the disallowed conformations are extreme limits of the allowed conformations? Today, with the availability of a large number of high resolution crystal structures, we have revisited this problem. Apart from validating some of the earlier findings such as residue propensities, preferred location in the secondary structure, we have explored their spatial neighborhood preferences using the protein structure network PSN] approach developed in our lab. Finally, the structural and functional implications of the disallowed conformations are examined.

PROPENSITIES OF AMINO ACID RESIDUES IN PROTEINS FOR DIFFERENT REGIONS OF THE RAMACHANDRAN MAP

Using a dataset of 1164 crystal structures of largely non-homologous proteins defined at a resolution of 1.5 angstrom or better, we have investigated the (phi,psi) preferences of 20 residue types by considering the residues which occur in loops. Propensities of residue types to occur in the loops with (phi,psi) values in the aa region of the Ramachandran map has a poor correlation coefficient of 0.48 to the Chou-Fasman propensities of the residue types to occur in the a-helical segments. However the correlation coefficient between propensities of residues in loops to adopt beta conformations and those in beta-sheet is much higher (0.95). These observations suggest that a-helix formation is well influenced by the local amino acid sequence while intrinsic preference of residue types for beta-sheet plays a major role in the formation of beta-sheet. The main chain polar groups of residues in loops, that can affect the (phi,psi) values, can be involved in intra-molecular hydrogen bonding. Therefore we investigated further by considering subset of residues in loops with low (0 to 2) number of intra-molecular hydrogen bonds per residue involving main chain polar atoms. For this subset, the correlation coefficients between propensities for alpha-helix and alpha(R) region and between beta-sheet and beta-region are 0.26 and 0.64 respectively. This reiterates higher intrinsic tendency of beta-region favouring residues to adopt beta-sheet than alpha(R) region favouring residues to adopt alpha-helical structure.

Mono- and Dialkyne Insertion Reactions of Cyclopalladated N,N `,N `'-Triarylguanidines kappa(2)(C,N)Pd(mu-Br)](2) and cis-/trans-kappa(2)(C,N)Pd(Lewis Base)Br]. Scaffolds for Enlarged, Rearranged, and Zwitterionic Palladacycles through Ring Contraction cum Amine-Imine Tautomerization

Insertion reactions of six-membered cyclopalladated N,N',N''-triarylguanidines, kappa(2)(C,N)Pd(mu-Br)](2) with various alkynes in CH2Cl2 under ambient conditions afforded diinserted eight-membered palladacycles, (kappa(2)(C,N):eta(2)(C=C)-PdBr] (1-11), in high yield (76-96%), while insertion reactions of six-membered cyclopalladated N,N',N''-triarylguanidines, kappa(2)(C,N)Pd(Lewis base)Br] (VI-XI), with various alkynes under the aforementioned conditions afforded monoinserted six-membered palladacycles, kappa(2)(C,N)-Pd(Lewis base)Br] (12-21), in high yield (81-91%) except for 14 (23%). The insertion reaction of VI with 2 equiv of dimethyl acetylenedicarboxylate (DMAD) and the insertion reaction of 12 with 1 equiv of DMAD in CH2Cl2 under ambient conditions resulted in the formation of a diinserted zwitterionic five-membered palladacycle, kappa(2)(C,C)Pd(2,6-lutidine)Br] (22), in 76% and 70% yields, respectively. Palladacycle 22 upon reaction with AgOTf in wet MeCN afforded the ionic palladacycle kappa(2)(C,C)Pd(2,6-lutidine)(H2O)]OTf] (23) in 78% yield. The ring size of the ``kappa(2)(C,N)Pd]'' unit in the structurally characterized diinserted palladacycles (1 center dot 2CH(2)Cl(2)center dot H2O, 2, 5, and 7), and monoinserted palladacycles (17, 18, and 20 center dot C7H8 H2O) is smaller than that anticipated for mono- and diinserted palladacycles, and this feature is mainly ascribed to the proclivity of III-XI to undergo ring contraction cum amine-imine tautomerization upon alkyne insertion. Palladacycle 22 represents the first diinserted product obtained in alkyne insertion reactions of kappa(2)(C,N)Pd(Lewis base)X] type palladarycles. The molecular structure of 22 center dot H2O determined by X-ray diffraction indicates that the positive charge on the guanidinium moiety is balanced by the negative charge on the palladium atom and thus represents the first structurally characterized zwitterionic palladacycle to be reported in alkyne insertion chemistry. Plausible mechanisms of formation of 12-21 and 22 have been outlined. The presence of more than one species in solution for some of the palladacycles in the series 1-7 and 12-21 was explained by invoking the C-N single-bond rotation of the CN3 unit of the guanidine moiety, while this process in conjunction with Pd-N(lutidine) bond rotation was invoked to explain the presence of four isomers of 15, as studied with the aid of variable-concentration H-1 NMR experiments carried out for 14 and 15.

HupB, a Nucleoid-Associated Protein of Mycobacterium tuberculosis, Is Modified by Serine/Threonine Protein Kinases In Vivo

HU, a widely conserved bacterial histone-like protein, regulates many genes, including those involved in stress response and virulence. Whereas ample data are available on HU-DNA communication, the knowledge on how HU perceives a signal and transmit it to DNA remains limited. In this study, we identify HupB, the HU homolog of the human pathogen Mycobacterium tuberculosis, as a component of serine/threonine protein kinase (STPK) signaling. HupB is extracted in its native state from the exponentially growing cells of M. tuberculosis H37Ra and is shown to be phosphorylated on both serine and threonine residues. The STPKs capable of modifying HupB are determined in vitro and the residues modified by the STPKs are identified for both in vivo and the in vitro proteins through mass spectrometry. Of the identified phosphosites, Thr(65) and Thr(74) in the DNA-embracing beta-strand of the N-terminal domain of HupB (N-HupB) are shown to be crucial for its interaction with DNA. In addition, Arg(55) is also identified as an important residue for N-HupB-DNA interaction. N-HupB is shown to have a diminished interaction with DNA after phosphorylation. Furthermore, hupB is shown to be maximally expressed during the stationary phase in M. tuberculosis H37Ra, while HupB kinases were found to be constitutively expressed (PknE and PknF) or most abundant during the exponential phase (PknB). In conclusion, HupB, a DNA-binding protein, with an ability to modulate chromatin structure is proposed to work in a growth-phase-dependent manner through its phosphorylation carried out by the mycobacterial STPKs.