Visualization and validation of (Q)SAR models

Analyzing and modeling relationships between the structure of chemical compounds, their physico-chemical properties, and biological or toxic effects in chemical datasets is a challenging task for scientific researchers in the field of cheminformatics. Therefore, (Q)SAR model validation is essential to ensure future model predictivity on unseen compounds. Proper validation is also one of the requirements of regulatory authorities in order to approve its use in real-world scenarios as an alternative testing method. However, at the same time, the question of how to validate a (Q)SAR model is still under discussion. In this work, we empirically compare a k-fold cross-validation with external test set validation. The introduced workflow allows to apply the built and validated models to large amounts of unseen data, and to compare the performance of the different validation approaches. Our experimental results indicate that cross-validation produces (Q)SAR models with higher predictivity than external test set validation and reduces the variance of the results. Statistical validation is important to evaluate the performance of (Q)SAR models, but does not support the user in better understanding the properties of the model or the underlying correlations. We present the 3D molecular viewer CheS-Mapper (Chemical Space Mapper) that arranges compounds in 3D space, such that their spatial proximity reflects their similarity. The user can indirectly determine similarity, by selecting which features to employ in the process. The tool can use and calculate different kinds of features, like structural fragments as well as quantitative chemical descriptors. Comprehensive functionalities including clustering, alignment of compounds according to their 3D structure, and feature highlighting aid the chemist to better understand patterns and regularities and relate the observations to established scientific knowledge. Even though visualization tools for analyzing (Q)SAR information in small molecule datasets exist, integrated visualization methods that allows for the investigation of model validation results are still lacking. We propose visual validation, as an approach for the graphical inspection of (Q)SAR model validation results. New functionalities in CheS-Mapper 2.0 facilitate the analysis of (Q)SAR information and allow the visual validation of (Q)SAR models. The tool enables the comparison of model predictions to the actual activity in feature space. Our approach reveals if the endpoint is modeled too specific or too generic and highlights common properties of misclassified compounds. Moreover, the researcher can use CheS-Mapper to inspect how the (Q)SAR model predicts activity cliffs. The CheS-Mapper software is freely available at http://ches-mapper.org.

Visuell-mediale Landschaftskonstruktionen als Bild-Raum-Diskurse

Durch die massenmediale Zunahme von statischen und bewegten Bilder im Laufe des letzten Jahrhunderts vollzieht sich unsere lebensweltliche Wirklichkeitskonstruktion zu Beginn des 21. Jahrhunderts zunehmend über Visualisierungen, die mit den neuen Formen der Digitalisierung noch an Dynamik zunehmen werden. Mit diesen omnipräsenten visuell-medialen Repräsentationen werden meist räumliche Vorstellungen transportiert, denn Räume werden vor allem über Bilder konstruiert. Diese Bildräume zirkulieren dabei nicht als singuläre Bedeutungszuschreibungen, sondern sind in sprachliche und bildliche Diskurse eingebettet. Visuell-mediale Bild-Raum-Diskurse besitzen zunehmend die Fähigkeit, unser Wissen über und unsere Wahrnehmung von Räumen zu kanalisieren und auf stereotype Raumstrukturen zu reduzieren. Dabei verfestigt sich eine normative Ordnung von bestimmten machtvollen Bildräumen, die nicht genügend kritisch hinterfragt werden. Deshalb ist es für die Geographie von entscheidender Wichtigkeit, mediale Raumkonstruktio- nen, ihre Einbettung in diskursive Bildarchive und ihre essentialistische und handlungspraktische gesellschaftliche Wirkung zu verstehen.rnLandschaften können vor diesem Hintergrund als visuell-medial transportierte Bild-Raum-Diskurse konzeptionalisiert werden, deren gesellschaftliche Wirkmächtigkeit mit Hilfe einer visuell ausgerichteten Diskursanalyse hinterfragt werden sollte. Auf Grundlage einer zeichentheoretischen Ikonologie wurde eine Methodik entwickelt, die visuell ausgerichtete Schrift-Bild-Räume angemessen analysieren kann. Am Beispiel der Inszenierung des Mittelrheintals, wurde, neben einer diachronischen Strukturanalyse der diskursrelevanten Medien (Belletristik, Malerei, Postkarten, Druckgrafiken und Fotografien), eine Feinanalyse der fotografischen „Rheinlandschaften“ von August Sander der 1930er Jahre durchgeführt. Als Ergebnis zeigte sich, dass der Landschaftsdiskurs über das Mittelrheintal immer noch durch die gegenseitige Durchdringung der romantischen Literatur und Malerei in der ersten Hälfte des 19. Jahrhunderts und die historischen Fotografien in den ersten Jahrzehnten des 20. Jahrhunderts bestimmt ist, nicht zuletzt forciert durch die Ernennung zum UNESCO-Welterbe 2002. Der stark visuell ausgerichtete Landschaftsdiskurs trägt somit zum einen positiv konnotierte, romantisch-pittoreske Züge, die die Einheit von Mensch und Natur symbolisieren, zum anderen historisch-konservatorische Züge, die eine Mythifizierung zu einer gewachsenen, authentischen Kulturlandschaft evozieren.

Applications of GIS and remote sensing techniques for studying the housing patterns in Cairo Governorate

The main objective of this study is to reveal the housing patterns in Cairo as one of the most rapidly urbanizing city in the developing world. The study outlines the evolution of the housing problem and its influencing factors in Egypt generally and in Cairo specifically. The study takes into account the political transition from the national state economy to the open door policy, the neo-liberal period and finally to the housing situation after the January 2011 Revolution. The resulting housing patterns in Cairo Governorate were identified as (1) squatter settlements, (2) semi-informal settlements, (3) deteriorated inner pockets, and (4) formal settlements. rnThe study concluded that the housing patterns in Cairo are reflecting a multifaceted problem resulting in: (1) the imbalance between the high demand for affordable housing units for low-income families and the oversupply of upper-income housing, (2) the vast expansion of informal areas both on agricultural and desert lands, (3) the deterioration of the old parts of Cairo without upgrading or appropriate replacement of the housing structure, and (4) the high vacancy rate of newly constructed apartmentsrnThe evolution and development of the current housing problem were attributed to a number of factors. These factors are demographic factors represented in the rapid growth of the population associated with urbanization under the dictates of poverty, and the progressive increase of the prices of both buildable land and building materials. The study underlined that the current pattern of population density in Cairo Governorate is a direct result of the current housing problems. Around the depopulation core of the city, a ring of relatively stable areas in terms of population density has developed. Population densification, at the expense of the depopulation core, is characterizing the peripheries of the city. The population density in relation to the built-up area was examined using Landsat-7 ETM+ image (176/039). The image was acquired on 24 August 2006 and considered as an ideal source for land cover classification in Cairo since it is compatible with the population census 2006.rnConsidering that the socio-economic setting is a driving force of change of housing demand and that it is an outcome of the accumulated housing problems, the socio-economic deprivations of the inhabitants of Cairo Governorate are analyzed. Small administrative units in Cairo are categorized into four classes based on the Socio-Economic Opportunity Index (SEOI). This index is developed by using multiple domains focusing on the economic, educational and health situation of the residential population. The results show four levels of deprivation which are consistent with the existing housing patterns. Informal areas on state owned land are included in the first category, namely, the “severely deprived” level. Ex-formal areas or deteriorated inner pockets are characterized as “deprived” urban quarters. Semi-informal areas on agricultural land concentrate in the third category of “medium deprived” settlements. Formal or planned areas are included mostly in the fourth category of the “less deprived” parts of Cairo Governorate. rnFor a better understanding of the differences and similarities among the various housing patterns, four areas based on the smallest administrative units of shiakhat were selected for a detailed study. These areas are: (1) El-Ma’desa is representing a severely deprived squatter settlement, (2) Ain el-Sira is an example for an ex-formal deprived area, (3) El-Marg el-Qibliya was selected as a typical semi-informal and medium deprived settlement, and (4) El-Nozha is representing a formal and less deprived area.rnThe analysis at shiakhat level reveals how the socio-economic characteristics and the unregulated urban growth are greatly reflected in the morphological characteristics of the housing patterns in terms of street network and types of residential buildings as well as types of housing tenure. It is also reflected in the functional characteristics in terms of land use mix and its degree of compatibility. It is concluded that the provision and accessibility to public services represents a performance measure of the dysfunctional structure dominating squatter and semi-informal settlements on one hand and ample public services and accessibility in formal areas on the other hand.rn