O VEGF é um fator mitogênico específico de células endoteliais que promove diferenciação celular materno-fetal placentária quando ligado a seus receptores (Flt-1 e KDR). Sua expressão é controlada por mecanismos autócrinos e parácrinos e está associada ao desenvolvimento da placenta. A placenta bovina foi utilizada como modelo de estudo por apresentar a facilidade de se avaliar os componentes do sistema VEGF em diferentes fases gestacionais. Como objetivo este estudo buscou analisar o fator de crescimento vascular endotelial (VEGF) e seus receptores através da técnica de PCR em tempo real no início, meio e fim de gestação. Para tanto, amostras de placentomas, região interplacentomal e corpo lúteo foram coletadas em diferentes fases gestacionais. Foram utilizados placentomas de animais clonados obtidos apenas aos 270 dias de gestação e estas amostras foram comparadas aos animais não clonados na mesma fase. A expressão do VEGF no placentoma apresentou um decréscimo (p < 0.05) no final da gestação (270 dias) em relação à expressão do VEGF aos 90 dias. A expressão do Flt-1 e do KDR na região interplacentomal foi semelhante desde os 45 até 90 dias de gestação e apresentou um aumento significativo (p < 0.05) aos 150 dias. No corpo lúteo gestacional, a expressão do VEGF aos 210 dias foi maior (p ≤ 0.05) em relação a 90 e 150 dias; observou-se também baixa expressão do KDR aos 90 dias de gestação (p < 0.05) em relação aos 210 dias. Pode-se concluir que a regulação da expressão do VEGF variou em relação aos seus receptores nos três tecidos avaliados. Placentomas de bovinos clonados não apresentaram diferenças significativas em relação à expressão do sistema VEGF se comparados aos placentomas de animais não clonados sugerindo ser esta expressão equivalente em placentas de animais clonados que vieram a termo. / The VEGF is a specific endothelial mitogenic factor that promotes feto-maternal cell differentiation in placenta through binding to its receptors (Flt-1 and KDR). Their expression is controlled by autocrine and paracrine mechanisms that are associated to placenta development. The bovine placenta was used in this study as a model due to easiness of evaluation of VEGF system components in different phases of pregnancy. The objective of this study was to analyze the vascular endothelial growth factor (VEGF) and its receptors expression using the real time PCR technique in the beginning, half and end of pregnancy. Furthermore, placentome samples, interplacentomal areas and corpus luteum were collected in different gestational phases for comparative studies. Placentome of cloned animals were analyzed at 270 days of pregnancy and compared to non-cloned animals in the same phase. The expression of VEGF in the placentome presented a decrease of expression (p < 0.05) in the end of the gestation (270 days) in relation to 90 days. The expression of Flt-1 and of KDR in interplacentomal area was similar from 45 to 90 days of pregnancy with a significant increase (p <0.05) observed at 150 days. In the gestational corpus luteum, the expression of VEGF at 210 days was higher (p ≤ 0.05) in comparison to 90 and 150 days. In the same tissue KDR expression at 90 days was lower (p < 0.05) in relation to 210 days. In conclusion the regulation VEGF varied in relation to its receptors expression in all three studied tissues. Cloned placentomes showed no significant differences in VEGF system expression compared to the placentome of non-cloned animals, suggesting there is an equivalent expression in placentas from cloned animals that came to term.

Piracicaba-SP é situada no extremo Sul da área preferencial para plantio de seringueira no Brasil, com pouca informação disponível sobre o desempenho de clones sob suas condições ambientais. O desempenho de crescimento e produtividade de dez clones de seringueira (Hevea spp.) foi avaliado. Os clones estudados foram CATI 21, IAC 15, IAN 873, GT 1, PB 235, PB 252, PR 107, PR 261, RRIM 526 e RRIM 600. O ensaio foi instalado no campo experimental do Departamento de Produção Vegetal da ESALQ/USP, em Piracicaba-SP, Brasil, sob delineamento de blocos inteiramente casualizados com cinco repetições. Foi avaliado o número e porcentagem de plantas em sangria e a produtividade anual. A sangria iniciou-se com a idade de 6,5 anos das árvores, no sistema de explotação de ½ S d/7 9m/y ET 3.3% 9/y. IAN 873 e PB 252 foram os maiores produtores, suplantando a produção do RRIM 600, que apresentou produtividade similar a do CATI 21 e do IAC 15. PR 107 e GT1 apresentaram produtividade intermediária, logo abaixo daqueles, mas com boas características secundárias como de produtividade e crescimento tardios e crescentes. PB 235, de forma inesperada, foi pouco produtivo. RRIM 526 teve produtividade imediatamente abaixo do PB 235 e sem qualquer característica secundária atrativa para recomendá-lo. PR 261 apresentou a produtividade mais baixa entre todos os clones. IAN 873, PB 252, RRIM 600, pelo bom desempenho no campo experimental e em diversas outras situações, são elegíveis para recomendação aos produtores para plantio comercial em grande escala, na região de Piracicaba. CATI 21 e IAC 15, pelo bom desempenho no campo experimental são elegíveis para recomendação aos produtores para plantio comercial em escala experimental, na região de Piracicaba. PR 261 não deve ser recomendado para plantio na região de Piracicaba. / Piracicaba-SP is on the extreme southern region suitable for rubber planting in Brazil, with little information on the performance of clones under its environmental conditions. The performance of growth and yield of ten rubber tree clones (Hevea spp.) were evaluated. The studied clones were CATI 21, IAC 15, IAN 873, GT 1, PB 235, PB, 252, PR 107, PR 261, RRIM 526 and RRIM 600. The trial was established in the experimental field of de Crop Production Department of the ESALQ/USP, at Piracicaba-SP, Brazil, under complete randomized block design with five replications. It was assessed the number and percentage of trees under tapping and yearly rubber yield. Tapping started at the trees age of 6,5 years, with the ½ S d/7 9m/y ET 3.3% 9/y exploitation system. IAN 873 and PB 252 were the highest producers, out-yielding RRIM 600, which presented similar yield then CATI 21 and IAC 15. PR 107 and GT1 had intermediate yield, immediately below the formers, but with good secondary characteristics as late and increasing growth and yield. PB 235, unexpectedly, was a low yielder. RRIM 526 was immediately below PB 235 without any attractive characteristic to recommend it. PR 261 presented the lowest yield. IAN 873, PB 252, RRIM 600, because their good performance in the experiment and their widely-tested good performance elsewhere, should be recommended for commercial planting in large scale. CATI 21 and IAC 15, because their good performance in the experiment, should be recommended for commercial planting in small scale. PR 261 should not be recommended for commercial planting in Piracicaba.

This study had the objective to evaluate mortality and growth of different eucalyptus clones under inoculation of mycorrhizal fungi as well as and mortality and growth of paricá under inoculation of mycorrhizal fungi and rhizobacteria in the southeast of Pará state, Brazil. The experiment with the eucalyptus clones was installed following a randomized block design with three replicates (three blocks) with 12 treatments and six clones. The seedling spacing was 3.0 x 3.0 meters with each plot having 56 seedlings. Half of the seedlings were inoculated with the arbuscular mycorrhizals fungi (endomycorrhizal) Glomus etunicatum, Glomus clarum, and the ectomycorrhizal Pisolithus microcarpus and the other half of the seedlings had no inoculation and served as a control. Each eucalyptus clone had 168 seedlings, totaling 1008 seedlings. Up to the last experiment measurement, the inoculation with the mycorrhizal fungi did not promote higher growth in diameter and height of the eucalyptus clones tested, when compared to the treatment without inoculation. The clones VM01 Eucalyptus urocan, 373 E. platyphylla, and 1250 E. urograndis had the lowest mortality rates. Therefore, the most recommended clones for Southeastern Pará were VM01 E. urocan, A 2017 E. urograndis, 373 E. platyphylla and 1250 E. urograndis because they presented better growth in diameter, height and lower mortality rate. The experiment with paricá was installed following the randomized block design with three replicates (three blocks) for each treatment being one replicate with 49 individuals, totaling 147 trees per treatment. The seedling spacing was 3.0 x 3.0 meters. In total, 588 paricá seedlings were used in the experiment. The development of paricá was evaluated under the influence of four treatments: (T0) control, (T1) conventional chemical fertilization (thermophosphate and, in the cover, NPK), (T2) inoculation with arbuscular mycorrhizal fungi, and (T3) inoculation with arbuscular mycorrhizal fungi and use of the plant growth promoting rhizobacteria. Inoculation with the plant growth promoter rhizobacteria (Enterobacter sp., Enterobacter aerogenes, Bacillus sp. and Pantoea sp.) and arbuscular mycorrhizal fungi in the species Glomus ethunicatum and Glomus clarum during the period evaluated did not influence development in diameter, height, and mortality of paricá plants. The paricá trees with conventional chemical fertilization had the best growth in height and diameter and lower mortality rate. However chemical fertilization is recommended in paricá plantations up four years in age in southeast Pará. / Este estudo teve como objetivo avaliar a mortalidade e o crescimento de diferentes clones de eucalipto sob inoculação de fungos micorrízicos e a mortalidade e crescimento de paricá sob inoculação de fungos micorrízicos e rizobactérias no sudeste do estado do Pará, Brasil. O experimento com os clones de eucalipto foi instalado seguindo um delineamento de blocos casualizados com três repetições (três blocos) com 12 tratamentos e seis clones. O espaçamento foi de 3,0 x 3,0 metros, com cada parcela tendo 56 mudas. Metade das plântulas foi inoculada com fungos micorrízicos arbusculares (endomicorrízicos) Glomus etunicatum, Glomus clarum e o ectomicorrízico Pisolithus microcarpus e a outra metade das plântulas não teve inoculação e serviu como controle. Até a última medição do experimento, a inoculação com os fungos micorrízicos não promoveu maior crescimento no diâmetro e altura dos clones de eucalipto testados, quando comparado ao tratamento sem inoculação. Os clones VM01 Eucalyptus urocan, 373 E. platyphylla e 1250 E. urograndis apresentaram as menores taxas de mortalidade. Portanto, os clones mais recomendados para o Sudeste do Pará foram VM01 E. urocan, A 2017 E. urograndis, 373 E. platyphylla e 1250 E. urograndis por apresentarem melhor crescimento em diâmetro, altura e menor taxa de mortalidade. O experimento com paricá foi instalado seguindo o delineamento de blocos casualizados com três repetições (três blocos) para cada tratamento, sendo cada repetição com 49 indivíduos, totalizando 147 árvores por tratamento. O espaçamento das mudas foi de 3,0 x 3,0 metros. O crescimento do paricá foi avaliado sob a influência de quatro tratamentos: (T0) controle, (T1) adubação química convencional (termofosfato e, na cobertura, NPK), (T2) inoculação com fungos micorrízicos arbusculares e (T3) inoculação com fungos micorrízicos arbusculares e uso de rizobactérias promotoras de crescimento de plantas. A inoculação com rizobactérias promotoras de crescimento (Enterobacter sp., Enterobacter aerogenes, Bacillus sp. e Pantoea sp.) e fungos micorrízicos arbusculares nas espécies Glomus ethunicatum e Glomus clarum durante o período avaliado não influenciou o crescimento em diâmetro, altura e mortalidade do paricá. As árvores de paricá com adubação química convencional apresentaram o melhor crescimento em altura e diâmetro e menor taxa de mortalidade. Entretanto, a adubação química é recomendada nas plantações de paricá com até quatro anos de idade no sudeste do Pará.

This research studied recombinant DNA-derived protein expression utilising expression vectors containing IRES sequences to link the gene of interest with the gene encoding selectable marker in mammalian cell cultures. Polycistronic expression constructs utilising internal ribosome entry site (IRES) can link unrelated genes under control of a single promoter. Transient study on the IRESlinked gene expression was performed. It was possible to standardise the level of protein expression to plasmid number by determining the number of free plasmids in the cytoplasm. The expression of a selectable marker when downstream of IRES was reduced in comparison to the monocistronic construct. Importantly when IRES was used, there were no negative effects on recombinant gene expression upstream of IRES. Down-regulating the selectable marker gene expression has been shown to enhance the probability of obtaining highly expressing clones. To investigate the effects of down-regulating fusion metallothionein green fluorescent protein (MTGFP), new constructs were created to combine metal inducible M2.6 promoter to drive the expression of human growth hormone linked to MTGFP by an attenuated IRES. This resulted in less MTGFP expression, reduced survivability and mean fluorescence in the presence of heavy metal. The increased metal sensitivity lengthened the initial selection period using reduced metal concentration in comparison to cells transfected with wildtype MTGFP. FACS can be used to select for resistance conferred by MTGFP despite reduced expression. FACS enrichment and sorting increased the hGH expression, which was correlated with mean fluorescence of the population; therefore fluorescence can be used as an indication of the final recombinant protein expression. Different approaches to isolate suitable clones were also investigated. It is preferable to select the transfected pool in low metal concentration for two weeks, sort for cells of high-fluorescence, and allow for recovery and proliferation. It is then possible to amplify gene expression by culturing the clones in increasing metal, resulting in further improvement of recombinant protein expression.

Pour la téléconférence, on peut remplacer l'image des correspondants distants par des modèles 3D animés de leurs visages. En plus de taux de compression avantageux, cette approche offre les libertés du virtuel : on peut par exemple composer à l'écran l'impression d'un lieu unique, virtuel, où débattent les représentants 3D. Cette thèse présente un algorithme de rendu spécifique, applicable à des clones 3D photo-réalistes de visages. En restreignant les angles de vue autorisés, il permet un rendu simple et rapide, même avec des ordinateurs peu puissants ou sur des machines virtuelles. On propose aussi une architecture de régie automatique, reliée à des caméras virtuelles qui réagissent aux interventions des participants et en proposent une image synthétique. En alternant plusieurs vues (éventuellement partielles) de la scène, on autorise plus de participants simultanés, sans compromettre la qualité de l'image proposée ni l'intelligibilité du débat restitué. Automatique, cette approche libère l'utilisateur du système, qui peut se concentrer sur un débat rendu plus attractif, à la fois comme spectateur et comme participant. Cette thèse rend aussi compte de la réalisation d'un prototype de communication qui intègre les éléments précédents et permet de juger la qualité de la communication obtenue. À cette occasion, l'utilisation d'un environnement sonore qui intègre les interventions distantes et leur localisation (dans ou hors de l'image) est discutée, avec plusieurs expérimentations sur l'association entre l'image et le son spatialisé. Enfin, on introduit une solution hybride (3D et vidéo) pour animer les clones des visages. En incrustant à la surface d'un clone statique l'image des yeux, des sourcils et de la bouche vues par une caméra, on laisse aux spectateurs la responsabilité d'interpréter les expressions originales, dans toute leur dimension vidéo (forte résolution spatiale et temporelle). Un second prototype permet de juger de l'empathie visuelle.

Homeobox genes encode homeodomain (HD) proteins which function as transcription factors and play an important role in plant and animal development by controlling cell specification and pattern formation. (Knotted1 in Arabidopsis thaliana) KNAT1-like mRNAs referred to as PtKN1(HD+) and mRNA sequences which lack HD region referred as PtKN1(hd-) were cloned from embryos of loblolly pine (Pinus taeda L.). Production of PtKN1(hd-) mRNAs is developmentally regulated and their encoded protein is abundant in mature pine embryos. Both forms of PtKN1 are produced by the same gene which has 5 exons; the regulatory dynamic is between cleavage-polyadenylation or termination within intron 3 to produce PtKN1 mRNA lacking HD sequences and splicing of exon 3 to exon 4 which excludes the 3'UTR/exon3 sequence to create an mRNA which encodes a HD. KNAT1 mRNA in Arabidopsis which lacks HD sequences was identified and characterized. While KNAT1 has been studied for many years, this is the first report of a KNAT1 mRNA lacking HD. KNAT1 mRNA lacking HD sequences was identified for the RS1 gene of maize, a monocotyledon. This is the first report of splicing of KNAT1 genes to produce mRNAs lacking HD sequences. The phenomenon appears to be ubiquitous as it is observed in gymnosperms, and both dicotyledonous and monocotyledonous angiosperms.

The methods presented in this study are aimed at the identification of subpopulations (clones) of genetically similar cells within tissue samples through measurement of loci-specific Fluorescence in-situ hybridization (FISH) spot signals for each nucleus and analyzing cell spatial distributions by way of Voronoi tessellation and Delaunay triangulation to robustly define cell neighbourhoods. The motivation for the system is to examine lung cancer patient for subpopulations of Non-Small Cell Lung Cancer (NSCLC) cells with biologically meaningful gene copy-number profiles: patterns of genetic alterations statistically associated with resistance to cis-platinum/vinorelbine doublet chemotherapy treatment. Current technologies for gene-copy number profiling rely on large amount of cellular material, which is not always available and suffers from limited sensitivity to only the most dominant clone in often heterogeneous samples. Thus, through the use of FISH, the detection of gene copy-numbers is possible in unprocessed tissues, allowing identification of specific tumour clones with biologically relevant patterns of genetic aberrations. The tissue-wide characterization of multiplexed loci-specific FISH signals, described herein, is achieved through a fully automated, multicolour fluorescence imaging microscope and object segmentation algorithms to identify cell nuclei and FISH spots within. Related tumour clones are identified through analysis of robustly defined cell neighbourhoods and cell-to-cell connections for regions of cells with homogenous and highly interconnected FISH spot signal characteristics. This study presents experiments which demonstrate the system’s ability to accurately quantify FISH spot signals in various tumour tissues and in up to 5 colours simultaneously or more through multiple rounds of FISH staining. Furthermore, the system’s FISH-based cell classification performance is evaluated at a sensitivity of 84% and specificity 81% and clonal identification algorithm results are determined to be comparable to clone delineation by a human-observer. Additionally, guidelines and procedures to perform anticipated, routine analysis experiments are established.

2004 – 2005 m. Alytaus AB „Vilda“ gamybinėse gluosnių ir karklų plantacijose įrengtose lauko kolekcijose buvo atlikta gluosnio žilvičio (S. viminalis L.) ir kai kurių jo kultivarų bei hibridų klonų (S. viminalis kl. 04116, S. viminalis `Americana` žaliažievė forma kl. 9972, S. viminalis `Americana` kl. 9976, S. viminalis `Tordis`, S. purpurea x S. viminalis kl. 9714 ir kl. 04141) morfobiologiniai ir produktyvumo tyrimai. Nustatyta, kad I – II auginimo metais tirtųjų taksonų krūmai skiriasi pagal jų aukštį, atžalinių ūglių ir stiebų skaičių krūme, atžalinių ūglių vidutinį ilgį bei jų skersmenį ir nulaibėjimo pobūdį, lapų formą ir dydį, medienos ir šerdies plotį ūglio skerspjūvyje. Daugiausiai atžalinių ūglių (vytelių) antraisiais auginimo metais išaugino S. viminalis `Americana` žaliažievė forma kl. 9972 (649,8 tūkst.vnt./ha) ir S. purpurea x S. viminalis kl. 9714 (609,9 tūkst.vnt./ha). Intensyviausiu atžalinių ūglių krūmų augimu I-II-aisiais auginimo metais išsiskyrė S. viminalis `Tordis`, S. viminalis `Americana` kl. 9976 ir S. viminalis `Americana` žaliažievė forma kl. 9972. Didžiausią krūmo atžalinių ūglių masę I-II-aisiais auginimo metais išaugino S. viminalis `Americana` žaliažievė forma kl. 9972 (2004 m. – 17,1 t/ha,2005 m. – 44,4 t/ha), S. viminalis `Americana` kl. 9976 (2004 m. – 14,6 t/ha, 2005 m. – 48,9 t/ha), S. viminalis `Tordis` (2004 m. – 13,5 t/ha, 2005 m. – 34,8 t/ha). Didžiausią vieno krūmo vidutinė dvimečių... [toliau žr. visą tekstą] / In 2004 – 2006, in plantation of Alytus join stock company “Vilda”, there were researched and valued morphological characteristics and productivity of basket willow (Salix viminalis L.), some of its cultivares and clones of hybrids (S. viminalis cl. 04116, S. viminalis `Americana` green bark form cl. 9972, S. viminalis `Americana` cl. 9976, S. viminalis `Tordis`, S. purpurea x S. viminalis cl. 9714 and cl. 04141). It has been proved some differences between S. viminalis L., its cultivares and clones of hybrids in number and mass of sprouts and stems on a bush, the height of bushes, their diametre lenght and yield of twigs. The most valuable taxa of clones regarding productivity are S. viminalis `Americana` cl. 9972 (649,8 thou.unit/ha) and S. purpurea x S. viminalis cl. 9714 (609,9 thou.unit/ha). S. viminalis `Tordis`, S. viminalis `Americana` cl. 9972, S. viminalis `Americana` cl.9976 are the most intensively growing bushes and have the longest sprouts. In the field collections the highest mass of regrowth sprouts and stems are found by the clone 9976 of S. viminalis `Americana`(1828,9 gram) and S. viminalis `Tordis` (1774,9 gram). It has been discovered that within the period of 3 years taxa of S. viminalis L. reach the stated mass of dry stems per one ha: S. viminalis cl. 04116 – 35,5 t/ha, S. viminalis `Americana` žaliažievė forma cl. 9972 – 44,2 t/ha, S. viminalis `Americana` cl. 9976 – 50,1 t/ha, S. viminalis `Tordis` – 49,6 t/ha.

Eucalyptus spp. and hybrids dominate the global plantation forestry industry, and vegetative propagation through cuttings is the preferred method for their commercial use. However, the cuttings of some species and hybrids show recalcitrance to rooting. The first aim of this study was to improve percentage rooting of three clones of E. grandis x E. nitens (Clones 1, 2 and 3) identified by a commercial nursery as having variable rooting abilities. The second was to relate their rooting responses as cuttings to their rooting responses in vitro. Minicuttings (3.5 – 4 cm in length) (hereafter referred to as cuttings) were subjected to commercial nursery propagation practices. Initial results revealed that in the absence of exogenous plant growth regulators (PGRs), soft (juvenile, thin diameter) cuttings survived (87 – 95%) and rooted (29 – 32%) significantly better than hard (mature, thick diameter) ones (62 – 71% survival and 2 – 8% rooting). This validated the use of soft cuttings by the nursery and all subsequent studies were conducted with soft cuttings. The other nursery practice of applying the commercial rooting powder Seradix 2 (3 g kgˉ¹ indole-3-butyric acid [IBA]) adversely affected the survival and subsequent rooting of cuttings of Clones 1 and 2. Ensuing studies investigated: 1) the effect of mode of IBA application (powder vs. liquid); 2) concentrations of Seradix (0, 0.5, 1, 2 and 3 g kgˉ¹ IBA), applied at initial placement of cuttings and two weeks later; and 3) the influence of season on the survival and subsequent rooting of cuttings. Results showed that regardless of the mode of application, IBA significantly reduced percentage survival and rooting in cuttings of Clones 1 and 2. The delayed application of Seradix, two weeks after cuttings were initially set, resulted in a higher percentage survival and rooting than when cuttings were supplied with Seradix at initial placement. Nevertheless, the best survival for Clones 1, 2 and 3 (95%, 99% and 71%, respectively) and rooting (83%, 64% and 47%, respectively) occurred in the absence of Seradix. In addition, the survival and rooting of cuttings were seasonally variable, with particularly low rooting during winter (e.g. for Clone 1, 32%) when compared with summer (e.g. for Clone 1, 83%). Shoots from all the clones were multiplied in vitro, followed by elongation on either of two media (E1= kinetin, α-naphthalene acetic acid [NAA] and IBA; E2 = kinetin and indole-3-acetic acid [IAA]), and then rooting on 0, 0.1 or 1.0 mg 1ˉ¹ IBA. The latter were selected to typify the range of Seradix concentrations used for the cuttings (i.e. no IBA, low and high IBA concentrations). For all three clones, shoots elongated on E1 or E2 displayed high survival (> 80%) but failed to root without IBA in the rooting medium. For Clones 1, 2 and 3 the best in vitro survival (80%, 100% and 100%, respectively) and rooting (40%, 75% and 40%, respectively) occurred when shoots were elongated on E2 and rooted on 0.1 mg 1ˉ¹ IBA. However, 1.0 mg 1ˉ¹ IBA in the rooting medium severely inhibited survival (0 – 50%), irrespective of the clone or the elongation treatment used. Overall, cuttings demonstrated the best survival and rooting in the absence of exogenous IBA, which suggested that sufficient endogenous auxin was present within the shoots for successful root induction. The application of exogenous IBA may have disrupted the cuttings’ endogenous PGR balance resulting in an inhibition of survival and rooting. In vitro shoots required a low concentration of IBA (0.1 mg 1ˉ¹) in order to counteract the antagonistic effect of cytokinins that were supplied during the multiplication and elongation culture stages, and promote rhizogenesis. Essentially, both cuttings and in vitro shoots demonstrated adverse survival and rooting responses when subjected to excessively high IBA concentrations. / Thesis (M.Sc.)-University of KwaZulu-Natal, Durban 2013.

The methods presented in this study are aimed at the identification of subpopulations (clones) of genetically similar cells within tissue samples through measurement of loci-specific Fluorescence in-situ hybridization (FISH) spot signals for each nucleus and analyzing cell spatial distributions by way of Voronoi tessellation and Delaunay triangulation to robustly define cell neighbourhoods. The motivation for the system is to examine lung cancer patient for subpopulations of Non-Small Cell Lung Cancer (NSCLC) cells with biologically meaningful gene copy-number profiles: patterns of genetic alterations statistically associated with resistance to cis-platinum/vinorelbine doublet chemotherapy treatment. Current technologies for gene-copy number profiling rely on large amount of cellular material, which is not always available and suffers from limited sensitivity to only the most dominant clone in often heterogeneous samples. Thus, through the use of FISH, the detection of gene copy-numbers is possible in unprocessed tissues, allowing identification of specific tumour clones with biologically relevant patterns of genetic aberrations. The tissue-wide characterization of multiplexed loci-specific FISH signals, described herein, is achieved through a fully automated, multicolour fluorescence imaging microscope and object segmentation algorithms to identify cell nuclei and FISH spots within. Related tumour clones are identified through analysis of robustly defined cell neighbourhoods and cell-to-cell connections for regions of cells with homogenous and highly interconnected FISH spot signal characteristics. This study presents experiments which demonstrate the system’s ability to accurately quantify FISH spot signals in various tumour tissues and in up to 5 colours simultaneously or more through multiple rounds of FISH staining. Furthermore, the system’s FISH-based cell classification performance is evaluated at a sensitivity of 84% and specificity 81% and clonal identification algorithm results are determined to be comparable to clone delineation by a human-observer. Additionally, guidelines and procedures to perform anticipated, routine analysis experiments are established.