Survey of Machine Learning Techniques in Drug Discovery.

BACKGROUND: Drug discovery, which is the process of discovering new candidate medications, is very important for pharmaceutical industries. At its current stage, discovering new drugs is still a very expensive and time-consuming process, requiring Phases I, II and III for clinical trials. Recently, machine learning techniques in Artificial Intelligence (AI), especially the deep learning techniques which allow a computational model to generate multiple layers, have been widely applied and achieved state-of-the-art performance in different fields, such as speech recognition, image classification, bioinformatics, etc. One very important application of these AI techniques is in the field of drug discovery. METHODS: We did a large-scale literature search on existing scientific websites (e.g, ScienceDirect, Arxiv) and startup companies to understand current status of machine learning techniques in drug discovery. RESULTS: Our experiments demonstrated that there are different patterns in machine learning fields and drug discovery fields. For example, keywords like prediction, brain, discovery, and treatment are usually in drug discovery fields. Also, the total number of papers published in drug discovery fields with machine learning techniques is increasing every year. CONCLUSION: The main focus of this survey is to understand the current status of machine learning techniques in the drug discovery field within both academic and industrial settings, and discuss its potential future applications. Several interesting patterns for machine learning techniques in drug discovery fields are discussed in this survey.

Immune ontogeny and engraftment receptivity in the sheep fetus.

OBJECTIVE: The biologic explanation for fetal receptivity to donor engraftment and subsequent long-term tolerance following transplantation early in gestation is not known. We investigated the role fetal immune ontogeny might play in fetal transplantation tolerance in sheep. METHODS: Engraftment of allogeneic and xenogeneic HSC was determined 60 days following transplantation at different time points in sheep fetal gestation. Parallel analysis of surface differentiation antigen expression on cells from lymphoid organs of timed gestational age fetal sheep was determined by flow cytometry using available reagents. RESULTS: An engraftment window was identified after day 52 gestation lasting until day 71 (term gestation: 145 days). This period was associated with the expression of the leukocyte common antigen CD45 on all cells in the thymus. Double-positive and single-positive CD4 and CD8 cells began appearing in the thymus just prior (day 45 gestation) to the beginning of the engraftment window, while single-positive CD4 or CD8 cells do not begin appearing in peripheral organs until late in the engraftment period, suggesting deletional mechanisms may be operative. In concert, surface IgM-positive cells express CD45 in the thymus at day 45, with a comparable delay in the appearance of IgM/CD45 cells in the periphery until late in the engraftment window. CONCLUSIONS: These findings support a central role for the thymus in multilineage immune cell maturation during the period of fetal transplantation receptivity. Further, they suggest that fetal engraftment receptivity is due to gestational age-dependent deletional tolerance.

Human embryonic stem cell-derived hematopoietic cells are capable of engrafting primary as well as secondary fetal sheep recipients.

The human/sheep xenograft model has proven valuable in assessing the in vivo hematopoietic activity of stem cells from a variety of fetal and postnatal human sources. CD34+/lineage- or CD34+/CD38- cells isolated from human embryonic stem cells (hESCs) differentiated on S17 feeder layer were transplanted by intraperitoneal injections into fetal sheep. Chimerism in primary transplants was established with polymerase chain reaction (PCR) and flow cytometry of bone marrow and peripheral blood samples. Whole bone marrow cells harvested from a primary recipient were transplanted into a secondary recipient. Chimerism was established as described before. This animal was stimulated with human GM-CSF, and an increase in human hematopoietic activity was noted by flow cytometry. Bone marrow aspirations cultured in methylcellulose generated colonies identified by PCR to be of human origin. We therefore conclude that hESCs are capable of generating hematopoietic cells that engraft primary recipients. These cells also fulfill the criteria for long-term engrafting hematopoietic stem cells as demonstrated by engraftment and differentiation in the secondary recipient.

Development of a web-based database to manage American College of Emergency Physicians/American Academy of Pediatrics Emergency Information Forms.

Children with special health care needs require special advanced planning for their unique emergencies. A Web site has been developed to allow secure Internet access to a database of Emergency Information Forms developed using the American College of Emergency Physicians/American Academy of Pediatrics format. The content and organization of the Web site, found at http://www.memscis.org, are described. A tour of the site is available. A set of XML data elements has been defined. Additional disaster preparedness elements have been added to the American College of Emergency Physicians/American Academy of Pediatrics Emergency Information Forms. The organization, security, and relationship of the site to electronic health records are described.

Interactions and trafficking of Andes and Sin Nombre Hantavirus glycoproteins G1 and G2.

This study was designed to investigate the trafficking of Andes virus (ANDV) and Sin Nombre virus (SNV) glycoproteins and to determine if ANDV or SNV glycoproteins G1 and G2 could be substituted for each other while still retaining normal trafficking. Trafficking of Hantaan virus (HNTV) and SNV glycoproteins has been studied and conflicting results were published regarding the Golgi targeting of G1 and G2 when expressed individually. The results reported in this manuscript suggest that both SNV and ANDV G1 and G2 expressed together, either from a single glycoprotein precursor (GPC) or from separate cDNAs, co-localize to the Golgi complex (GC). When expressed individually, neither G1 nor G2 was able to translocate from the endoplasmic reticulum (ER) to the GC. Interestingly, when ANDV G1 and SNV G2 or ANDV G2 and SNV G1 are co-expressed, they interact and are colocalized in the GC.