Structure-activity relationships derived by machine learning: the use of atoms and their bond connectivities to predict mutagenicity by inductive logic programming.

We present a general approach to forming structure-activity relationships (SARs). This approach is based on representing chemical structure by atoms and their bond connectivities in combination with the inductive logic programming (ILP) algorithm PROGOL. Existing SAR methods describe chemical structure by using attributes which are general properties of an object. It is not possible to map chemical structure directly to attribute-based descriptions, as such descriptions have no internal organization. A more natural and general way to describe chemical structure is to use a relational description, where the internal construction of the description maps that of the object described. Our atom and bond connectivities representation is a relational description. ILP algorithms can form SARs with relational descriptions. We have tested the relational approach by investigating the SARs of 230 aromatic and heteroaromatic nitro compounds. These compounds had been split previously into two subsets, 188 compounds that were amenable to regression and 42 that were not. For the 188 compounds, a SAR was found that was as accurate as the best statistical or neural network-generated SARs. The PROGOL SAR has the advantages that it did not need the use of any indicator variables handcrafted by an expert, and the generated rules were easily comprehensible. For the 42 compounds, PROGOL formed a SAR that was significantly (P < 0.025) more accurate than linear regression, quadratic regression, and back-propagation. This SAR is based on an automatically generated structural alert for mutagenicity.

Use of an oocyte expression assay to reconstitute inductive signaling.

We have developed a paracrine signaling assay capable of mimicking inductive events in the early vertebrate embryo. RNA encoding one or more secreted proteins is microinjected into a Xenopus laevis oocyte. After a brief incubation to allow translation, a piece of embryonic tissue competent to respond to the signaling protein is grafted onto the oocyte. The secreted protein's effect on the grafted explant is then scored by assaying expression of tissue-specific markers. Explants of ectodermal tissue from blastula or gastrula stage embryos were grafted onto oocytes that had been injected with RNA encoding activin or noggin. We found that the paracrine assay faithfully reconstitutes mesoderm induction by activin and neural induction by noggin. Blastula-stage explants grafted onto activin-expressing oocytes expressed the mesodermal marker genes brachyury, goosecoid, and muscle actin. Gastrula-stage explants grafted onto noggin-expressing oocytes expressed neural cell adhesion molecule (NCAM) and formed cement gland. By injecting pools of RNA synthesized from a cDNA expression library into the oocyte, we also used the assay to screen for secreted neural-inducing proteins. We assayed 20,000 independent transformants of a library constructed from LiCl-dorsalized Xenopus laevis embryos, and we identified two cDNAs that induced neural tissue in ectodermal explants from gastrula-stage embryos. Both cDNAs encode noggin. These results suggest that the paracrine assay will be useful for the cloning of novel signaling proteins as well as for the analysis of known factors.

Basic fibroblast growth factor can mediate the early inductive events in renal development.

The earliest characterized events during induction of tubulogenesis in renal anlage include the condensation or compaction of metanephrogenic mesenchyme with the concurrent upregulation of WT1, the gene encoding the Wilms tumor transcriptional activator/suppressor. We report that basic fibroblast growth factor (FGF2) can mimic the early effects of an inductor tissue by promoting the condensation of mesenchyme and inhibiting the tissue degeneration associated with the absence of an inductor tissue. By in situ hybridization, FGF2 was also found to mediate the transcriptional activation of WT1 and of the hepatocyte growth factor receptor gene, c-met. Although FGF2 can induce these early events of renal tubulogenesis, it cannot promote the epithelial conversion associated with tubule formation in metanephrogenic mesenchyme. For this, an undefined factor(s) from pituitary extract in combination with FGF2 can cause tubule formation in uninduced mesenchyme. These findings support the concept that induction in kidney is a multiphasic process that is mediated by more than a single comprehensive inductive factor and that soluble molecules can mimic these inductive activities in isolated uninduced metanephrogenic mesenchyme.

What We Don't Know Can Hurt Us: Clinical Inference Bias and Constructed Memories of Sexual Abuse.

The false memory/recovered memory debate, research regarding the malleability of memory, and the current lack of methods for validating recovered memories all support the view that heightened care is required of therapists dealing with clients whom they suspect have been sexually abused. The judgmental heuristics that underlie the major clinical inference biases of confirmatory bias, biased covariation, base rate fallacies, and schematic processing errors are all relevant to the processes leading to therapist-client constructions of memories of sexual abuse. Suggestions for minimizing each of these biases are offered. Personal motivations of the client and client suggestibility are factors that may contribute to the construction of memories of sexual abuse, and suggestions for minimizing the impact of these motivations are offered. In conclusion, general suggestions for minimizing the impact of clinical inference biases within the sexual abuse treatment context are summarized.

Inference of binding affinity from neuronal receptors in humans

Tese de mestrado, Bioinformática e Biologia Computacional (Bioinformática), Universidade de Lisboa, Faculdade de Ciências, 2016

Intracranial EEG correlates of implicit relational inference within the hippocampus

Drawing inferences from past experiences enables adaptive behavior in future situations. Inference has been shown to depend on hippocampal processes. Usually, inference is considered a deliberate and effortful mental act which happens during retrieval, and requires the focus of our awareness. Recent fMRI studies hint at the possibility that some forms of hippocampus-dependent inference can also occur during encoding and possibly also outside of awareness. Here, we sought to further explore the feasibility of hippocampal implicit inference, and specifically address the temporal evolution of implicit inference using intracranial EEG. Presurgical epilepsy patients with hippocampal depth electrodes viewed a sequence of word pairs, and judged the semantic fit between two words in each pair. Some of the word pairs entailed a common word (e.g.,‘winter - red’, ‘red - cat’) such that an indirect relation was established in following word pairs (e.g, ‘winter - cat’). The behavioral results suggested that drawing inference implicitly from past experience is feasible because indirect relations seemed to foster ‘fit’ judgments while the absence of indirect relations fostered 'do not fit' judgments, even though the participants were unaware of the indirect relations. A event-related potential (ERP) difference emerging 400 ms post-stimulus was evident in the hippocampus during encoding, suggesting that indirect relations were already established automatically during encoding of the overlapping word pairs. Further ERP differences emerged later post-stimulus (1500 ms), were modulated by the participants' responses and were evident during encoding and test. Furthermore, response-locked ERP effects were evident at test. These ERP effects could hence be a correlate of the interaction of implicit memory with decision-making. Together, the data map out a time-course in which the hippocampus automatically integrates memories from discrete but related episodes to implicitly influence future decision making.

Modern statistical methods; descriptive and inductive,

Mode of access: Internet.

Inference rules for unsatisfiability /

UIUCDCS-R-79-955

A comparative discussion of variable-valued logic and grammatical inference /

Bibliography: p. 45-46.

The uniclass inductive program AQ7UNI : program implementation and user's guide /

"UILU-ENG 79 1729."

Learning without negative examples via variable-valued logic characterizations : the uniclass inductive program AQ7UNI /

"July 1979."

G-R-S automatic train control, intermittent inductive auto-manual system: operation, maintenance, adjustment.

Mode of access: Internet.

Principles of scientific botany; or, Botany as an inductive science.

Mode of access: Internet.

An English grammar, in familiar conversations, inductive and progressive; uniting and harmonizing theory and practice, and adapted to oral teaching.

Includes bibliographical references.