Machine Learning in Healthcare

Recent advancements in Artificial Intelligence (AI) and Machine Learning (ML) technology have brought on substantial strides in predicting and identifying health emergencies, disease populations, and disease state and immune response, amongst a few. Although, skepticism remains regarding the practical application and interpretation of results from ML-based approaches in healthcare settings, the inclusion of these approaches is increasing at a rapid pace. Here we provide a brief overview of machine learning-based approaches and learning algorithms including supervised, unsupervised, and reinforcement learning along with examples. Second, we discuss the application of ML in several healthcare fields, including radiology, genetics, electronic health records, and neuroimaging. We also briefly discuss the risks and challenges of ML application to healthcare such as system privacy and ethical concerns and provide suggestions for future applications.     

工业蛋白质理性设计与应用

作为合成生物学与绿色生物制造等领域的底层核心技术,蛋白理性设计可有效解决天然功能元件性能不足等共性挑战,创制高性能人工酶元件。值此天津工业生物研究所(Tianjin Institute of Industrial Biotechnology, TIB)创立10周年之际,文中回顾了研究所在工业蛋白理性设计领域的系列重要工作进展。从酶设计方法学研究、新酶反应设计到生物催化应用等方面进行了分析讨论,并展望了本领域未来发展方向。望借此搭建学术界和产业界与酶理性设计的桥梁,促进新技术、新策略的开发应用,加速融合人工酶的基础研究与产业应用,推动我国生物制造领域的科技创新升级。     

Genome physical mapping with large-insert bacterial clones by fingerprint analysis: methodologies, source clone genome coverage, and contig map quality

Genome physical mapping with large-insert clones by fingerprint analysis is becoming an active area of genomics research. Here, we report two new capillary electrophoresis-based fingerprinting methods for genome physical mapping and the effects of different fingerprinting methods and source clone genome coverage on quality physical map construction revealed by computer simulations and laboratory experiments. It was shown that the manual sequencing gel-based two-enzyme fingerprinting method consistently generated larger and more accurate contigs, followed by the new capillary electrophoresis-based three-enzyme method, the new capillary electrophoresis-based five-enzyme (SNaPshot) method, the agarose gel-based one-enzyme method, and the automatic sequencing gel-based four-enzyme method, in descending order, when 1% or fewer questionable clones were allowed. Analysis of clones equivalent to 5x, 8x, 10x, and 15x genomes using the fingerprinting methods revealed that as the number of clones increased from 5x to 10x, the contig length rapidly increased for all methods. However, when the number of clones was increased from 10x to 15x coverage, the contig length at best increased at a lower rate or even decreased. The results will provide useful knowledge and strategies for effective construction of quality genome physical maps for advanced genomics research.     

Solid-Phase Artificial Chaperone-Assisted Refolding using Insoluble β-Cyclodextrin–Acrylamide Copolymer Beads

Solid-phase refolding methods are advantageous since they facilitate both separation of solid additives from the refolded protein and recycling of the additives. -Cyclodextrin–acrylamide copolymer hydrogel beads were used as a matrix for detergents in solid-phase artificial chaperone-assisted refolding and improved the yield of lysozyme (up to 65%) and carbonic anhydrase B (up to 80%), compared with conventional solid host matrices.Revisions received 29 September 2004     

Insight into Trichoderma reesei's genome content, organization and evolution revealed through BAC library characterization

Trichoderma reesei is an important industrial fungus known for its ability to efficiently secrete large quantities of protein as well as its wide variety of biomass degrading enzymes. Past research on this fungus has primarily focused on extending its protein production capabilities, leaving the structure of its 33 Mb genome essentially a mystery. To begin to address these deficiencies and further our knowledge of T. reesei's secretion and cellulolytic potential, we have created a genomic framework for this fungus. We constructed a BAC library containing 9216 clones with an average insert size of 125 kb which provides a coverage of 28 genome equivalents. BAC ends were sequenced and annotated using publicly available software which identified a number of genes not seen in previously sequenced EST datasets. Little evidence was found for repetitive sequence in T. reesei with the exception of several copies of an element with similarity to the Podospora anserina transposon, PAT. Hybridization of 34 genes involved in biomass degradation revealed five groups of co-located genes in the genome. BAC clones were fingerprinted and analyzed using fingerprinted contigs (FPC) software resulting in 334 contigs covering 28 megabases of the genome. The assembly of these FPC contigs was verified by congruence with hybridization results.     

Dorsal telencephalon-specific expression of Cre recombinase in PAC transgenic mice

The ability to restrict gene expression or disruption to specific regions of the brain would enhance understanding of the molecular basis for brain development and function. For this purpose, brain region-restricted promoters are essential. Here we report the isolation of a DNA fragment containing the Emx1 gene promoter, which is responsible for dorsal telencephalon-specific expression. The Cre recombinase gene was inserted into a mouse PAC (P1-derived artificial chromosome) Emx1-locus clone (PAC-Emx1#1 clone) and utilized to generate three transgenic mouse lines. In all three lines, especially Tg3, Cre-mediated recombination was highly restricted to Emx1-expressing cell lineages, from embryonic stages to adulthood. Immunohistochemical analyses showed that Cre protein is expressed in the dorsal telencephalon in all three lines in adulthood. Thus, the PAC-Emx1#1 clone contains essentially all regulatory elements necessary for Emx1 gene expression. Our results suggest that Emx1-Cre Tg3 mice and the PAC-Emx1#1 clone constitute powerful tools for dorsal telencephalon-specific gene manipulation.     

间断性人工重力作用对模拟失重大鼠股骨的防护效应

目的：比较三种间断性人工重力对抗措施对模拟失重大鼠的影响。方法：SD雄性大鼠30只,按体重配对后随机等分为5组：对照组（C）,悬吊组（S）,站立组（G1）,1.5G（G2）和2.6G人工重力组（G3）。对抗组大鼠每日分别给予1h的站立、1.5和2.6G人工重力作用。利用物理测量和三点弯曲等实验,观察了3周间断性人工重力对大鼠股骨生长、生物力学特性的影响。结果：与S组大鼠比较：G1组弹性载荷、最大和刚性系数显著恢复（P＜0.05）;G2组直径（P＜0.01）和干重、密度（P＜0.05）,弹性载荷和最大载荷显著提高（P＜0.05）;G3组弹性载荷和最大载荷显著恢复（P＜0.01）。结论：三种对抗措施均显著改善了尾吊大鼠承重骨的生长力学特性,而1.5G的1h/d人工重力作用是较理想的对抗方法。     

On the role of periodism in the origin of proteins

Two different views have been proposed for origins of genes (or proteins). One is that primordial genes evolved from random sequences. This view underlies the concept of modern in vitro evolution experiments that functional molecules (even proteins) evolved from random sequence-libraries. On the contrary, the second view reminds that "random sequences" would be an unusual state in which to find RNA or DNA, because it is their inherent nature to yield periodic structures during the course of semi-conservative replication. In this second view, the periodicity of DNA (or RNA) is responsible for emergence of primordial genes. Although recent reports on the variety of periodicities present in proteins, genes and genomes are consistent with the second view, it has yet to be experimentally tested. We assessed the significance of periodicities of DNA in the origin of genes by constructing such periodic DNAs. The results showed that periodic DNA produced ordered proteins at very high rates, which is in contrast to the fact that proteins with random sequences lack secondary structures. We concluded that periodicity played a pivotal role in the origin of many genes. The observation should pave the way for new experimental evolution systems for proteins.     

The Immune Response to Artificial Proteins Containing Biologically Active Fragments of Human IFN-α2 and Insulin

A study was made of the humoral immune response of BALB/c mice to various doses of artificial proteins that contained biologically active fragments of human interferon 2 (IFN-2) and insulin. The insulin fragment had no effect on the response to any protein construct. The IFN-2 fragment increased the titer of antibodies against the construct. Mapping of continuous B epitopes with immune sera revealed several antigenic determinants, the C end of the IFN-2 fragment with the adjacent de novo protein region being immunodominant. The more effective binding of serum antibodies with the constructs containing the IFN-2 fragment was attributed to antibody interaction with the fragment and to better recognition of the entire protein construct by the immune system.     

AFLP-derived SCARs facilitate construction of a 1.1 Mb sequence-ready map of a region that spans the Vf locus in the apple genome

The availability of high-density anchored markers is a prerequisite for reliable construction of a deep coverage BAC contig, which leads to creation of a sequence-ready map in the target chromosomal region. Unfortunately, such markers are not available for most plant species, including woody perennial plants. Here, we report on an efficient approach to build a megabase-size sequence-ready map in the apple genome for the Vf region containing apple scab resistance gene(s) by targeting AFLP-derived SCAR markers to this specific genomic region. A total of 11 AFLP-derived SCAR markers, previously tagged to the Vf locus, along with three other Vf-linked SCAR markers have been used to screen two apple genome BAC libraries. A single BAC contig which spans the Vf region at a physical distance of approximately 1,100 kb has been constructed by assembling the recovered BAC clones, followed by closure of inter-contig gaps. The contig is 4 ×deep, and provides a minimal tiling path of 16 contiguous and overlapping BAC clones, thus generating a sequence-ready map. Within the Vf region, duplication events have occurred frequently, and the Vf locus is restricted to the ca. 290 kb region covered by a minimum of three overlapping BAC clones.