Accurate clinical toxicity prediction using multi-task deep neural nets and contrastive molecular explanations.

Explainable machine learning for molecular toxicity prediction is a promising approach for efficient drug development and chemical safety. A predictive ML model of toxicity can reduce experimental cost and time while mitigating ethical concerns by significantly reducing animal and clinical testing. Herein, we use a deep learning framework for simultaneously modeling in vitro, in vivo, and clinical toxicity data. Two different molecular input representations are used; Morgan fingerprints and pre-trained SMILES embeddings. A multi-task deep learning model accurately predicts toxicity for all endpoints, including clinical, as indicated by the area under the Receiver Operator Characteristic curve and balanced accuracy. In particular, pre-trained molecular SMILES embeddings as input to the multi-task model improved clinical toxicity predictions compared to existing models in MoleculeNet benchmark. Additionally, our multitask approach is comprehensive in the sense that it is comparable to state-of-the-art approaches for specific endpoints in in vitro, in vivo and clinical platforms. Through both the multi-task model and transfer learning, we were able to indicate the minimal need of in vivo data for clinical toxicity predictions. To provide confidence and explain the model's predictions, we adapt a post-hoc contrastive explanation method that returns pertinent positive and negative features, which correspond well to known mutagenic and reactive toxicophores, such as unsubstituted bonded heteroatoms, aromatic amines, and Michael receptors. Furthermore, toxicophore recovery by pertinent feature analysis captures more of the in vitro (53%) and in vivo (56%), rather than of the clinical (8%), endpoints, and indeed uncovers a preference in known toxicophore data towards in vitro and in vivo experimental data. To our knowledge, this is the first contrastive explanation, using both present and absent substructures, for predictions of clinical and in vivo molecular toxicity.

Advances in neuroscience imply that harmful experiments in dogs are unethical.

Functional MRI (fMRI) of fully awake and unrestrained dog 'volunteers' has been proven an effective tool to understand the neural circuitry and functioning of the canine brain. Although every dog owner would vouch that dogs are perceptive, cognitive, intuitive and capable of positive emotions/empathy, as indeed substantiated by ethological studies for some time, neurological investigations now corroborate this. These studies show that there exists a striking similarity between dogs and humans in the functioning of the caudate nucleus (associated with pleasure and emotion), and dogs experience positive emotions, empathic-like responses and demonstrate human bonding which, some scientists claim, may be at least comparable with human children. There exists an area analogous to the 'voice area' in the canine brain, enabling dogs to comprehend and respond to emotional cues/valence in human voices, and evidence of a region in the temporal cortex of dogs involved in the processing of faces, as also observed in humans and monkeys. We therefore contend that using dogs in invasive and/or harmful research, and toxicity testing, cannot be ethically justifiable.

Ahimsa and alternatives -- the concept of the 4th R. The CPCSEA in India.

The Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA) in India is one of a kind in the world. It is a statutory body of the government of India formed by an act of the Indian parliament. This body consists of nominated members and representatives from national regulatory agencies, Ministry of Health and Family Welfare, Ministry of Environment and Forests, national academic and research councils, premier research institutes, eminent scientists and animal welfare organisations. The CPCSEA draws its powers from the Prevention of Cruelty to Animals (PCA) Act of 1960 which states that the duty of the committee is "to take all such measures as may be necessary to ensure that animals are not subject to unnecessary pain or suffering before, during or after the performance of experiments on them". With the power to promulgate its own laws to ensure the humane and ethical use of animals in research and education, the CPCSEA in 1998 notified in the gazette of India the "Breeding of and Experiments on Animals (Control and Supervision) Rules 1998". The CPCSEA is unique in that the law in itself has enabled the creation of a common platform of discussion for scientists and animal activists for humane and progressive solutions for the use of animals in experimentation. In a country that is caught in a paradox of violence and rich cultural and religious traditions, India still draws a lot of its power from the concept of "Ahimsa" (the philosophy of non-violence). This concept is also pertinent to the use of animals in laboratories. Unethical, inhumane and unscientific practices, and ignorance of the use of alternatives were the way of science until 1999 when CPCSEA became functional. For four years CPCSEA has waged a battle, rescued thousands of animals from laboratories, fought legal battles to victory, enforced for the first time in the country good laboratory practice, designed guidelines for the use of animals in the production of immunobiologicals, introduced the credo of 3R principles, trained and taught scientific personnel the credibility of humane science and most importantly brought forward the concept of the fourth R, "rehabilitation" of used laboratory animals. Today CPCSEA has made it a national policy that personnel using experimental animals have a moral responsibility towards these animals after their use. Costs of after-care/rehabilitation of animals post experimentation are to be a part of research costs and should be scaled in positive correlation with the level of sentience of the animals. This paper is about the Indian law on animal experimentation and the success story of the CPCSEA in India in inculcating the credo of 4Rs -- Replacement, Reduction, Refinement, and Rehabilitation of animals used in experimentation.

Animal experimentation and ethics in India: the CPCSEA makes a difference.

The Committee for the Purpose of Control and Supervision of Experiments on Animals (CPCSEA) is a statutory body formed by the Act of the Indian Parliament under the Prevention of Cruelty to Animals Act 1960. Formed in 1964, it was revived in 1998, under the committed chairpersonship of Maneka Gandhi. In the last two years, the CPCSEA has bettered the life of the animals in laboratories across India. This committee is composed of members of the scientific community, regulatory authorities and animal activists. The CPCSEA functions with a brilliant network of volunteers who liaise with the laboratories. For the first time in India: over 665 laboratories are registered with the CPCSEA; Institutional Animal Ethics Committees (IAECs) are constituted in every laboratory, which are only empowered to approve research project proposals that use rats, mice, guinea-pigs or rabbits; every project that uses canines, ovines, bovines or non-human primates can only be conducted if approved by the panel of scientific experts constituted for this purpose; guidelines on laboratory animal care and practice have been formulated and enforced; a protocol for the production of immunobiologicals from equines has been formulated and ratified by the Supreme Court of India; the CPCSEA has been deliberating on alternatives and working out modalities to introduce alternatives in basic/regulatory research and education, in keeping with the international arena; the CPCSEA, to date, has rehabilitated and homed over 300 dogs, 150 equines, 200 non-human primates and several cattle, cats, birds, rabbits and mice; the CPCSEA proactively trains and guides scientific and non-scientific personnel on issues of alternatives and laboratory animal welfare; and the CPCSEA has fought legal issues on laboratory animal care and use and have had verdicts that favoured alternatives and animal welfare.