Pathformer: a biological pathway informed transformer for disease diagnosis and prognosis using multi-omics data.

MOTIVATION: Multi-omics data provide a comprehensive view of gene regulation at multiple levels, which is helpful in achieving accurate diagnosis of complex diseases like cancer. However, conventional integration methods rarely utilize prior biological knowledge and lack interpretability. RESULTS: To integrate various multi-omics data of tissue and liquid biopsies for disease diagnosis and prognosis, we developed a biological pathway informed Transformer, Pathformer. It embeds multi-omics input with a compacted multi-modal vector and a pathway-based sparse neural network. Pathformer also leverages criss-cross attention mechanism to capture the crosstalk between different pathways and modalities. We first benchmarked Pathformer with 18 comparable methods on multiple cancer datasets, where Pathformer outperformed all the other methods, with an average improvement of 6.3%-14.7% in F1 score for cancer survival prediction, 5.1%-12% for cancer stage prediction, and 8.1%-13.6% for cancer drug response prediction. Subsequently, for cancer prognosis prediction based on tissue multi-omics data, we used a case study to demonstrate the biological interpretability of Pathformer by identifying key pathways and their biological crosstalk. Then, for cancer early diagnosis based on liquid biopsy data, we used plasma and platelet datasets to demonstrate Pathformer's potential of clinical applications in cancer screening. Moreover, we revealed deregulation of interesting pathways (e.g. scavenger receptor pathway) and their crosstalk in cancer patients' blood, providing potential candidate targets for cancer microenvironment study. AVAILABILITY AND IMPLEMENTATION: Pathformer is implemented and freely available at https://github.com/lulab/Pathformer.

Generation of the salicylic acid deficient Arabidopsis via a synthetic salicylic acid hydroxylase expression cassette.

BACKGROUND: Salicylic acid (SA) is one of the plant hormones, which plays crucial roles in signaling transduction in plant growth, disease resistance, and leaf senescence. Arabidopsis (Arabidopsis thaliana) SA 3-hydroxylase (S3H) and 5-hydroxylase (S5H) are key enzymes which maintain SA homeostasis by catalyzing SA to 2,3-dihydroxybenzoic acid (DHBA) and 2,5-DHBA, respectively. RESULTS: SA deficient transgenic Arabidopsis lines were generated by introducing two binary vectors S5Hpro::EGFP-S3H and 35Spro::EGFP-S3H respectively, in which the expression of S3H is under the control of the S5H promoter or CaMV 35S promoter. Compared with the constitutive expression of S3H gene under the control of 35S promoter, the S3H gene under the native S5H promoter is activated by endogenous SA and results in a dynamic control of SA catabolism in a feedback mode. The SA accumulation, growth, leaf senescence, and pathogen resistance of the S5Hpro::GFP-S3H transgenic plants were investigated in parallel with NahG transgenic plants. The SA levels in the S5Hpro::EGFP-S3H transgenic plants were similar to or slightly lower than those of NahG transgenic Arabidopsis and resulted in SA deficient phenotypes. The low-SA trait of the S5Hpro::EGFP-S3H transgenic lines was inherited stably in the later generations. CONCLUSIONS: Compared with NahG transgenic lines producing by-product catechol, S5Hpro::EGFP-S3H transgenic lines reduce SA levels by converting SA to a native product 2,3-DHBA for catabolism. Together, we provide new SA-deficient germplasms for the investigations of SA signaling in plant development, leaf senescence, and disease resistance.

SENP1 in the retrosplenial agranular cortex regulates core autistic-like symptoms in mice.

Autism spectrum disorder (ASD) is a highly heritable neurodevelopmental disorder, causing defects of social interaction and repetitive behaviors. Here, we identify a de novo heterozygous gene-truncating mutation of the Sentrin-specific peptidase1 (SENP1) gene in people with ASD without neurodevelopmental delay. We find that Senp1+/- mice exhibit core autistic-like symptoms such as social deficits and repetitive behaviors but normal learning and memory ability. Moreover, we find that inhibitory and excitatory synaptic functions are severely affected in the retrosplenial agranular (RSA) cortex of Senp1+/- mice. Lack of Senp1 leads to increased SUMOylation and degradation of fragile X mental retardation protein (FMRP), also implicated in syndromic ASD. Importantly, re-introducing SENP1 or FMRP specifically in RSA fully rescues the defects of synaptic function and autistic-like symptoms of Senp1+/- mice. Together, these results demonstrate that disruption of the SENP1-FMRP regulatory axis in the RSA causes autistic symptoms, providing a candidate region for ASD pathophysiology.

Label-free detection of Giardia lamblia cysts using a deep learning-enabled portable imaging flow cytometer.

We report a field-portable and cost-effective imaging flow cytometer that uses deep learning and holography to accurately detect Giardia lamblia cysts in water samples at a volumetric throughput of 100 mL h-1. This flow cytometer uses lens free color holographic imaging to capture and reconstruct phase and intensity images of microscopic objects in a continuously flowing sample, and automatically identifies Giardia lamblia cysts in real-time without the use of any labels or fluorophores. The imaging flow cytometer is housed in an environmentally-sealed enclosure with dimensions of 19 cm × 19 cm × 16 cm and weighs 1.6 kg. We demonstrate that this portable imaging flow cytometer coupled to a laptop computer can detect and quantify, in real-time, low levels of Giardia contamination (e.g., <10 cysts per 50 mL) in both freshwater and seawater samples. The field-portable and label-free nature of this method has the potential to allow rapid and automated screening of drinking water supplies in resource limited settings in order to detect waterborne parasites and monitor the integrity of the filters used for water treatment.

Direct synthesis of the organic and Ge free Al containing BOG zeolite (ITQ-47) and its application for transformation of biomass derived molecules.

Aluminosilicate boggsite (Si/Al-BOG) has been hydrothermally synthesized without adding organic structure-directing agents (OSDAs) in the synthesis gel using the borosilicogermanium ITQ-47 (Si/B-ITQ-47) zeolite as seeds. The introduction of the costly and environmentally less benign phosphazene organic structure-directing agent is not required to grow the zeolite. Physicochemical characterization experiments show that Si/Al-BOG has good crystallinity, high surface area, tetrahedral Al3+ species, and acid sites. In order to test the catalytic performance of the zeolite, the synthesis of l,l-lactide from l-lactic acid was performed. Si/Al-BOG exhibits 88.2% conversion of l-lactic acid and 83.8% l,l-lactide selectivity, which are better than those of other zeolites studied up to now.

Correction: Direct synthesis of the organic and Ge free Al containing BOG zeolite (ITQ-47) and its application for transformation of biomass derived molecules.

[This corrects the article DOI: 10.1039/D0SC04044D.].