carlos sosa - Academia.edu (original) (raw)

Papers by carlos sosa

Journal of Medical Virology, 1992

Human parvovirus B19 productively infected erythroid progenitor (EP) cells from umbilical cord bl... more Human parvovirus B19 productively infected erythroid progenitor (EP) cells from umbilical cord blood, in vitro as shown by an increase of viral DNA in supernatant fluid assayed by dot blot hybridization and liquid scintillation counting. Progeny virus was released into the supernatant fluid of CD34+ EP cells which had been purified by immunomagnetic separation. This supernatant fluid was infectious for bone marrow cells. Erythroid bursts infected with virus showed characteristic cytopathic effect by electron microscopy consisting of cytoplasmic vacuolization, marginated chromatin, and nuclear inclusions of lattice or crystalline arrays. Cultures of umbilical cord blood EP cells may be useful for the propagation of parvovirus 619 serological testing reagents and the study of virus-host cell interactions.

The Journal of Physical Chemistry, 1989

+ NO2 reaction in the production of SO2 in chambers. The formation of MSA, CH3S03H, could also fo... more + NO2 reaction in the production of SO2 in chambers. The formation of MSA, CH3S03H, could also follow CH3S02 pro-d~c t i o n .~~ However, under atmospheric conditions of low NO2 different considerations may apply, and an oxidation chain initiated by CH3S + O2 could give a different product distribution. These questions can only be resolved by a thorough study of CH3S0 and CH3S02 chemistry and more mechanistic information on the CH3S + O2 reaction. Acknowledgment. This work was supported by NOAA as part of the National Acid Precipitation Assessment Program. We are grateful to D. Fahey for analyzing the N O mixture, A. Wahner for writing the data handling programs, and J. Smith for help with the design and construction of the detection optics. (46) Saltzman, E. S.; Savoie, D. L.; Zika, R. G.; Prospero, J. M.

Chemical Physics Letters, 1988

The reactions of fluorine atom and hydroxyl radical with molecular hydrogen have been studied by ... more The reactions of fluorine atom and hydroxyl radical with molecular hydrogen have been studied by molecular orbital methods using the 6-3 11 G (d, p) and 6-3 1 1 + f G (2df, 2pd) basis sets with the unrestricted Hartree-Fock approach and with Moller-Plesset perturbation theory up to fourth order, with and without spin projection. The positions oftransition states were optimized with a grid search at the MPn and PMPn levels using the 6-3 1 IG(d, p) basis. The projected MP2, MP3 and MP4 barriers are 1-I .5

International Journal of Quantum Chemistry, 1986

Heats of reaction and barrier heights have been computed for H + CH2CH2 → C2H5, H + CH2O → CH3O, ... more Heats of reaction and barrier heights have been computed for H + CH2CH2 → C2H5, H + CH2O → CH3O, and H + CH2O → CH2OH using unrestricted Hartree-Fock and Møller–Plesset perturbation theory up to fourth order (with and without spin annihilation), using single-reference configuration interaction, and using multiconfiguration self-consistent field methods with 3-21G, 6-31G(d), 6-31G(d,p), and 6-311G(d,p) basis sets. The barrier height in all three reactions appears to be relatively insensitive to the basis sets, but the heats of reaction are affected by p-type polarization functions on hydrogen. Computation of the harmonic vibrational frequencies and infrared intensities with two sets of polarization functions on heavy atoms [6-31G(2d)] improves the agreement with experiment. The experimental barrier height for H + C2H4 (2.04 ± 0.08 kcal/mol) is overestimated by 7‒9 kcal/mol at the MP2, MP3, and MP4 levels. MCSCF and CISD calculations lower the barrier height by approximately 4 kcal/mol relative to the MP4 calculations but are still almost 4 kcal/mol too high compared to experiment. Annihilation of the largest spin contaminant lowers the MP4SDTQ computed barrier height by 8‒9 kcal/mol. For the hydrogen addition to formaldehyde, the same trends are observed. The overestimation of the barrier height with Møller-Plesset perdicted barrier heights for H + C2H4 → C2H5, H + CH2O → CH3O, and H + CH2O → CH2OH at the MP4SDTQ/6-31G(d) after spin annihilation are respectively 1.8, 4.6, and 10.5 kcal/mol.

Journal of The American Chemical Society, 1987

The reactions of O H radical with C2H2 and CzH4 have been studied with ab initio molecular orbita... more The reactions of O H radical with C2H2 and CzH4 have been studied with ab initio molecular orbital techniques. Reactants, loose clusters, transition structures, and products were optimized at UHF/3-21G and UHF/6-3 1G*. The barrier heights have been computed by using unrestricted Hartree-Fock and Mdler-Plesset perturbation theory up to fourth order, including single, double, and quadruple excitations. Spin contamination in the UHF wave function has been corrected by annihilating the largest spin contaminant. The vibrational frequencies were computed by using analytical derivative methods at the UHF/3-21G level. The barrier heights for both reactions are overestimated by 7-15 kcal/mol at the UMPZ, UMP3, and UMP4 levels. Annihilation of the largest spin contaminant lowers the barrier heights by 7-15 kcal/mol. Calculations at the PMP4/6-3 lG* level are in good agreement with the estimated experiment barrier heights.

Journal of Reproductive Immunology, 1998

Kaposi's sarcoma (KS) is a form of skin cancer, most commonly found in individuals suffering from... more Kaposi's sarcoma (KS) is a form of skin cancer, most commonly found in individuals suffering from acquired immunodeficiency syndrome, or AIDS. However, before the worldwide infection of human immunodeficiency virus (HIV), the rare occurrence of KS was confined to two distinct groups of individuals. In the Western world, the classical form of KS was often found in older men (60-70 years of age) from the Mediterranean area. Another form called endemic KS, was found in Equatorial Africa. Currently, the most common cases of KS are found in individuals suffering from AIDS. This is called AIDS-associated KS. Between 30 and 40% of male, homosexual AIDS patients suffer from AIDS-associated KS. KS is also occasionally diagnosed in transplant patients receiving immunosuppressive drugs (to keep their body from rejecting the foreign organ). As opposed to cases of classic and endemic KS, the KS in AIDS patients progresses very quickly, often with a fatal outcome. Human herpesvirus type 8 (HHV-8) has been implicated as the cause of Kaposi's sarcoma (KS), but the exact connection of the virus to the neoplasm is not known. The virus has been detected within the sarcoma skin lesions, but has additionally been seen in peripheral blood

have ranged widely. This is possibly due to differences in assay sensitivity, geographic or popul... more have ranged widely. This is possibly due to differences in assay sensitivity, geographic or population-based differences in the true presence of the virus in semen, and PCR contamination. This study assessed interlaboratory sensitivity and reproducibility in the analysis of blinded experimental panels, each consisting of 48 specimens and being composed of semen specimens from different healthy artificial-insemination donors (n ‫؍‬ 30) and human immunodeficiency virus (HIV)-infected patients (n ‫؍‬ 7) plus positive (n ‫؍‬ 4) and negative (n ‫؍‬ 7) controls. The experimental panels analyzed in each laboratory were identical except for being independently coded. Of 10 experiments done in five laboratories, 5 experiments from three laboratories had evidence of PCR contamination; all instances of contamination were in the context of nested PCR procedures. In the experiments with no false-positive results, HHV-8 DNA was detected in three (8%) of the 37 semen specimens (two from artificial-insemination donors and one from an HIV-positive patient) but in only 3 (1.6%) of the 184 PCRs in which these specimens were analyzed. This suggests that HHV-8 DNA is present in semen at concentrations that can be too low to allow its consistent detection. This study emphasizes the importance of performing blinded, multi-institution experiments to provide a coherent basis for comparing results and to motivate standardization of methods.

This paper presents our first experiences in mapping and optimizing genomic sequence search onto ... more This paper presents our first experiences in mapping and optimizing genomic sequence search onto the massively parallel IBM Blue Gene/P (BG/P) platform. Specifically, we performed our work on mpiBLAST, a parallel sequence-search code that has been optimized on numerous supercomputing environments. In doing so, we identify several critical performance issues. Consequently, we propose and study different approaches for mapping sequence-search and parallel I/O tasks on such massively parallel architectures. We demonstrate that our optimizations can deliver nearly linear scaling (93% efficiency) on up to 32,768 cores of BG/P. In addition, we show that such scalability enables us to complete a large-scale bioinformatics problem -sequence searching a microbial genome database against itself to support the discovery of missing genes in genomes -in only a few hours on BG/P. Previously, this problem was viewed as computationally intractable in practice.

Bayesian phylogenetic inference is an important alternative to maximum likelihood-based phylogene... more Bayesian phylogenetic inference is an important alternative to maximum likelihood-based phylogenetic method. However, inferring large trees using the Bayesian approach is computationally demanding-requiring huge amounts of memory and months of computational time. With a combination of novel parallel algorithms and latest system technology, terascale phylogenetic tools will provide biologist the computational power necessary to conduct experiments on very large dataset, and thus aid construction of the tree of life.

This IBM® Redpaper publication presents a virtual screening study of the DOCK Version 6.0 molecul... more This IBM® Redpaper publication presents a virtual screening study of the DOCK Version 6.0 molecular docking software package on a massively parallel system, the IBM System Blue Gene® supercomputer, Blue Gene/L. 1 Virtual screening of very large libraries of small ligands requires not only efficient algorithms but an efficient implementation for docking thousands, if not millions, of compounds simultaneously in a reasonable amount of time.

IEEE Transactions on Parallel and Distributed Systems, 2008

Bioinformatics databases used for sequence comparison and sequence alignment are growing exponen... more Bioinformatics databases used for sequence comparison and sequence alignment are growing exponentially. This has popularized programs that carry out database searches. Current implementations of sequence alignment methods based on hidden Markov models (HMM) have ...

In the life sciences, genomic databases for sequence search have been growing exponentially in si... more In the life sciences, genomic databases for sequence search have been growing exponentially in size. As a result, faster sequencesearch algorithms to search these databases continue to evolve to cope with algorithmic time complexity. The ubiquitous tool for such search is the Basic Local Alignment Search Tool (BLAST) [1] from the National Center for Biotechnology Information (NCBI). Despite continued algorithmic improvements in BLAST, it cannot keep up with the rate at which the database is exponentially increasing in size. Therefore, parallel implementations such as mpiBLAST have emerged to address this problem. The performance of such implementations depends on a myriad of factors including algorithmic, architectural, and mapping of the algorithm to the architecture. This paper describes modifications and extensions to a parallel and distributed-memory version of BLAST called mpiBLAST-PIO and how it maps to a massively parallel system, specifically IBM Blue Gene/L (BG/L). The extensions include a virtual file manager, a "multiple master" runtime model, efficient fragment distribution, and intelligent load balancing. In this study, we have shown that our optimized mpiBLAST-PIO on BG/L using a query with 28014 sequences and the NR and NT databases scales to 8192 nodes (two cores per node). The cases tested here are well suited for a massively parallel system.