Extreme Learning Machine (ELM) is a universal approximation method that is extremely fast and easy to implement, but the weights of the model are normally randomly selected so they can lead to poor prediction performance. In this work, we applied Weighted Similarity Extreme Learning Machine in combination with Jaccard/Tanimoto (WELM-JT) and cluster analysis (namely, k-means clustering and Support Vector Clustering) on similarity and distance measures (i.e., Jaccard/Tanimoto and Euclidean) in order to predict which compounds with not-so-different chemical structures have an activity for treating a certain symptom or disease. The proposed method was experimented on one of the most challenging datasets named Maximum Unbiased Validation (MUV) dataset with 4 different types of fingerprints (i.e. ECFP_4, ECFP_6, FCFP_4 and FCFP_6). The experimental results show that WELM-JT in combination with k-means-ED gave the best performance. It retrieved the highest number of active molecules and used the lowest number of nodes. Meanwhile, WELM-JT with k-means-JT and ECFP_6 encoding proved to be a robust contender for most of the activity classes.

Rapid opiate detoxification (ROD) is a technique whereby the opiate-dependent patient is withdrawn acutely, under anesthesia, from the opioid. Following detoxification, patients experience severe back pain and restlessness often accompanied by a restless-leg-syndrome. We evaluated gabapentin given immediately following detoxification to attenuate these symptoms. In addition, we evaluated the use of the somatosensory-evoked potential (SEP) as a parameter to quantitate pain responses. Patients (n = 21; mean age 32.5 +/- 7 SD; 12 males, 9 females) underwent ROD with naltrexone (2 x 50 mg) during propofol anesthesia and artificial ventilation (IPPV). Sympathetic overshoot was attenuated by clonidine, and increased bowl. movement was managed by continuous i.v. somatostatin. Back pain, restlessness, and restless-leg-syndrome were treated with gabapentin (1200 mg) in the ICU. Efficacy was assessed by the patient's subjective ratings of restlessness (0-4). In addition, measurements of amplitude (muV), latency (ms) of late N-100-peak of the somatosensory evoked potential (SEP), and tolerance to an increased electrical nociceptive stimulus (mA) to the forearm were performed. Data were compared to pre-treatment control and to the period shortly after detoxification. From a mean of 8.4 +/- 2.5 muV, N-100-peak increased to a mean of 12.3 muV +/- 3.3 (p < 0.005) following opioid detoxification. Gabapentin reduced amplitude height to a mean of 3.5 +/- 1.5 muV. Also, tolerance to nociceptive stimulus, which had dropped to 4.4 mA, increased to 12.5 mA (p < 0.01), while intensity for restlessness and thrashing of limbs dropped from 3.2 to 1.2 (p < 0.05). The sudden displacement of the opiate from its receptor site induced by naltrexone, resulted in a post inhibitory SEP overshoot with an increase in nociceptive afferent volleys, and a lowering in pain threshold. This was associated with back pain, limb thrashing and a restless-leg-syndrome, all of which could be attenuated by gabapentin. The amplitude of late N-100-peak parameter appears to be a potential candidate to quantify the increase of nociception in such patients. (C) 2004 Published by Elsevier SAS.