The potential of Raman-activated cell sorting (RACS) is inherently limited by conflicting demands for signal quality and sorting throughput. Here, we present positive dielectrophoresis–based Raman-activated droplet sorting (pDEP-RADS), where a periodical pDEP force was exerted to trap fast-moving cells, followed by simultaneous microdroplet encapsulation and sorting. Screening of yeasts for triacylglycerol (TAG) content demonstrated near-theoretical-limit accuracy, ~120 cells min⁻¹ throughput and full-vitality preservation, while sorting fatty acid degree of unsaturation (FA-DU) featured ~82% accuracy at ~40 cells min⁻¹. From a yeast library expressing algal diacylglycerol acyltransferases (DGATs), a pDEP-RADS run revealed all reported TAG-synthetic variants and distinguished FA-DUs of enzyme products. Furthermore, two previously unknown DGATs producing low levels of monounsaturated fatty acid–rich TAG were discovered. This first demonstration of RACS for enzyme discovery represents hundred-fold saving in time consumables and labor versus culture-based approaches. The ability to automatically flow-sort resonance Raman–independent phenotypes greatly expands RACS’ application.

Multi-Task Learning (MTL) is designed to train multiple correlated tasks simultaneously, thereby enhancing the performance of individual tasks. Typically, a multi-task network structure consists of a shared backbone and task-specific decoders. However, the complexity of the decoders increases with the number of tasks. To tackle this challenge, we integrate the decoder-free vision-language model CLIP, which exhibits robust zero-shot generalization capability. Recently, parameter-efficient transfer learning methods have been extensively explored with CLIP for adapting to downstream tasks, where prompt tuning showcases strong potential. Nevertheless, these methods solely fine-tune a single modality (text or visual), disrupting the modality structure of CLIP. In this paper, we first propose Multi-modal Alignment Prompt (MmAP) for CLIP, which aligns text and visual modalities during fine-tuning process. Building upon MmAP, we develop an innovative multi-task prompt learning framework. On the one hand, to maximize the complementarity of tasks with high similarity, we utilize a gradient-driven task grouping method that partitions tasks into several disjoint groups and assign a group-shared MmAP to each group. On the other hand, to preserve the unique characteristics of each task, we assign an task-specific MmAP to each task. Comprehensive experiments on two large multi-task learning datasets demonstrate that our method achieves significant performance improvements compared to full fine-tuning while only utilizing approximately ~ 0.09% of trainable parameters.

Large-scale pre-trained models have achieved remarkable success in various computer vision tasks. A standard approach to leverage these models is to fine-tune all model parameters for downstream tasks, which poses challenges in terms of computational and storage costs. Recently, inspired by Natural Language Processing (NLP), parameter-efficient transfer learning has been successfully applied to vision tasks. However, most existing techniques primarily focus on single-task adaptation, and despite limited research on multi-task adaptation, these methods often exhibit suboptimal training/inference efficiency. In this paper, we first propose an once-for-all Vision Multi-Task Adapter (VMT-Adapter), which strikes approximately O(1) training and inference efficiency w.r.t task number. Concretely, VMT-Adapter shares the knowledge from multiple tasks to enhance cross-task interaction while preserves task-specific knowledge via independent knowledge extraction modules. Notably, since task-specific modules require few parameters, VMT-Adapter can handle an arbitrary number of tasks with a negligible increase of trainable parameters. We also propose VMT-Adapter-Lite, which further reduces the trainable parameters by learning shared parameters between down- and up-projections. Extensive experiments on four dense scene understanding tasks demonstrate the superiority of VMT-Adapter(-Lite), achieving a 3.96% (1.34%) relative improvement compared to single-task full fine-tuning, while utilizing merely ～1% (0.36%) trainable parameters of the pre-trained model.

Microalgae are promising feedstock for production of lipids, sugars, bioactive compounds and in particular biofuels, yet development of sensitive and reliable phylotyping strategies for microalgae has been hindered by the paucity of phylogenetically closely-related finished genomes. Using the oleaginous eustigmatophyte Nannochloropsis as a model, we assessed current intragenus phylotyping strategies by producing the complete plastid (pt) and mitochondrial (mt) genomes of seven strains from six Nannochloropsis species. Genes on the pt and mt genomes have been highly conserved in content, size and order, strongly negatively selected and evolving at a rate 33% and 66% of nuclear genomes respectively. Pt genome diversification was driven by asymmetric evolution of two inverted repeats (IRa and IRb): psbV and clpC in IRb are highly conserved whereas their counterparts in IRa exhibit three lineage-associated types of structural polymorphism via duplication or disruption of whole or partial genes. In the mt genomes, however, a single evolution hotspot varies in copy-number of a 3.5 Kb-long, cox1-harboring repeat. The organelle markers (e.g., cox1, cox2, psbA, rbcL and rrn16_mt) and nuclear markers (e.g., ITS2 and 18S) that are widely used for phylogenetic analysis obtained a divergent phylogeny for the seven strains, largely due to low SNP density. A new strategy for intragenus phylotyping of microalgae was thus proposed that includes (i) twelve sequence markers that are of higher sensitivity than ITS2 for interspecies phylogenetic analysis, (ii) multi-locus sequence typing based on rps11_mt-nad4, rps3_mt and cox2-rrn16_mt for intraspecies phylogenetic reconstruction and (iii) several SSR loci for identification of strains within a given species. This first comprehensive dataset of organelle genomes for a microalgal genus enabled exhaustive assessment and searches of all candidate phylogenetic markers on the organelle genomes. A new strategy for intragenus phylotyping of microalgae was proposed which might be generally applicable to other microalgal genera and should serve as a valuable tool in the expanding algal biotechnology industry.

Microalgae can convert carbon dioxide into organic matter through photosynthesis. Thus, they are considered as an environment-friendly and efficient cell chassis for biologically active metabolites. Microalgal lipids are a class of organic compounds that can be used as raw materials for food, feed, cosmetics, healthcare products, bioenergy, etc., with tremendous potential for commercialization. In this review, we summarized the commercial lipid products from eukaryotic microalgae, and updated the mechanisms of lipid synthesis in microalgae. Moreover, we reviewed the enhancement of lipids, triglycerides, polyunsaturated fatty acids, pigments, and terpenes in microalgae via environmental induction and/or metabolic engineering in the past five years. Collectively, we provided a comprehensive overview of the products, biosynthesis, induced strategies and genetic engineering in microalgal lipids. Meanwhile, the outlook has been presented for the development of microalgal lipids industries, emphasizing the significance of the accurate analysis of lipid bioactivity, as well as the high-throughput screening of microalgae with specific lipids.

Since the discovery of cell-free foetal DNA in the plasma of pregnant women, many non-invasive prenatal testing assays have been developed. In the area of skeletal dysplasia diagnosis, some PCR-based non-invasive prenatal testing assays have been developed to facilitate the ultrasound diagnosis of skeletal dysplasias that are caused by de novo mutations. However, skeletal dysplasias are a group of heterogeneous genetic diseases, the PCR-based method is hard to detect multiple gene or loci simultaneously, and the diagnosis rate is highly dependent on the accuracy of the ultrasound diagnosis. In this study, we investigated the feasibility of using targeted capture sequencing to detect foetal de novo pathogenic mutations responsible for skeletal dysplasia. Three families whose foetuses were affected by skeletal dysplasia and two control families whose foetuses were affected by other single gene diseases were included in this study. Sixteen genes related to some common lethal skeletal dysplasias were selected for analysis, and probes were designed to capture the coding regions of these genes. Targeted capture sequencing was performed on the maternal plasma DNA, the maternal genomic DNA, and the paternal genomic DNA. The de novo pathogenic variants in the plasma DNA data were identified using a bioinformatical process developed for low frequency mutation detection and a strict variant interpretation strategy. The causal variants could be specifically identified in the plasma, and the results were identical to those obtained by sequencing amniotic fluid samples. Furthermore, a mean of 97% foetal specific alleles, which are alleles that are not shared by maternal genomic DNA and amniotic fluid DNA, were identified successfully in plasma samples. Our study shows that capture sequencing of maternal plasma DNA can be used to non-invasive detection of de novo pathogenic variants. This method has the potential to be used to facilitate the prenatal diagnosis of skeletal dysplasia.