Predicting the potential distribution of invasive alien pests (i.e. habitat suitability modelling) and their potential spread from existing populations (i.e. habitat susceptibility modelling) is critical to guide management responses at local, regional and national scales. We use the management of Chilean needle grass (Nassella neesiana) invasion in a 260,791 km2 part of eastern Australia as an example to describe a process-based approach for making such predictions with publicly available soft ware (e.g. Netica and ESRI products). The approach is deductive, with causal relationships captured in a Bayesian network and represented spatially at fine resolution using a geographic information system (GIS). Pest risk responses to changing environments, such as land-use change, climate change or altered flood regimes, and to management interventions can be tested through scenario analysis. Predictive risk mapping of invasive aliens is often knowledge-constrained; therefore, our approach seeks to capture the best available knowledge from often disparate sources in a transparent and explicit manner. For Chilean needle grass, we elicited process understanding from experts through a participatory approach, integrated an existing bioclimatic model and obtained our own field data. Our model, thereby, represents a hypothesis of what determines the distribution, abundance and spread of Chilean needle grass in the modelled region. Specifically, the model forecasts the likelihood of the weed reaching a threshold density (e.g. in this case, >30% ground cover) as defined by the experts. This approach to likelihood estimation contrasts with the presence/absence predictions of most other models. Modelling was done at a sufficiently fine spatial resolution (i.e. 30 m) to capture relevant invasion dynamics. Finally, we illustrate how validation can be used to give end users confidence in model predictions and to identify important knowledge gaps and uncertainties. We demonstrate how the resulting pest risk maps for Chilean needle grass can guide management decisions.

We present fast methods for filtering voltage measurements and performing optimal inference of the location and strength of synaptic connections in large dendritic trees. Given noisy, subsampled voltage observations we develop fast l 1 -penalized regression methods for Kalman state-space models of the neuron voltage dynamics. The value of the l 1 -penalty parameter is chosen using crossvalidation or, for low signal-to-noise ratio, a Mallows' C p -like criterion. Using low-rank approximations, we reduce the inference runtime from cubic to linear in the number of dendritic compartments. We also present an alternative, fully Bayesian approach to the inference problem using a spike-andslab prior. We illustrate our results with simulations on toy and real neuronal geometries. We consider observation schemes that either scan the dendritic geometry uniformly or measure linear combinations of voltages across several locations with random coefficients. For the latter, we show how to choose the coefficients to offset the correlation between successive measurements imposed by the neuron dynamics. This results in a "compressed sensing" observation scheme, with an important reduction in the number of measurements required to infer the synaptic weights.

We investigate Bayesian methods for optimal decoding of noisy or incompletely-observed spike trains. Information about neural identity or temporal resolution may be lost during spike detection and sorting, or spike times measured near the soma may be corrupted with noise due to stochastic membrane channel effects in the axon. We focus on neural encoding models in which the (discrete) neural state evolves according to stimulus-dependent Markovian dynamics. Such models are sufficiently flexible that we may incorporate realistic stimulus encoding and spiking dynamics, but nonetheless permit exact computation via efficient hidden Markov model forward-backward methods. We analyze two types of signal degradation. First, we quantify the information lost due to jitter or downsampling in the spike-times. Second, we quantify the information lost when knowledge of the identities of different spiking neurons is corrupted. In each case the methods introduced here make it possible to quantify the dependence of the information loss on biophysical parameters such as firing rate, spike jitter amplitude, spike observation noise, etc. In particular, decoders that model the probability distribution of spike-neuron assignments significantly outperform decoders that use only the most likely spike assignments, and are ignorant of the posterior spike assignment uncertainty.

No abstract