CodonBERT: a BERT-based architecture tailored for codon optimization using the cross-attention mechanism.

MOTIVATION: Due to the varying delivery methods of mRNA vaccines, codon optimization plays a critical role in vaccine design to improve the stability and expression of proteins in specific tissues. Considering the many-to-one relationship between synonymous codons and amino acids, the number of mRNA sequences encoding the same amino acid sequence could be enormous. Finding stable and highly expressed mRNA sequences from the vast sequence space using in silico methods can generally be viewed as a path-search problem or a machine translation problem. However, current deep learning-based methods inspired by machine translation may have some limitations, such as recurrent neural networks, which have a weak ability to capture the long-term dependencies of codon preferences. RESULTS: We develop a BERT-based architecture that uses the cross-attention mechanism for codon optimization. In CodonBERT, the codon sequence is randomly masked with each codon serving as a key and a value. In the meantime, the amino acid sequence is used as the query. CodonBERT was trained on high-expression transcripts from Human Protein Atlas mixed with different proportions of high codon adaptation index codon sequences. The result showed that CodonBERT can effectively capture the long-term dependencies between codons and amino acids, suggesting that it can be used as a customized training framework for specific optimization targets. AVAILABILITY AND IMPLEMENTATION: CodonBERT is freely available on https://github.com/FPPGroup/CodonBERT.

[Risk factors for intra-abdominal hypertension in children with sepsis].

OBJECTIVE: To study risk factors for the occurrence of intra-abdominal hypertension (IAH) in children with sepsis. METHODS: A nest case-control study was employed. According to intra-abdominal pressures (IAP) measured by cystometry, 119 children with sepsis were classified into normal IAP (control, n = 80) and IAH groups (n = 39). Risk factors for the occurrence of IAH were investigated by monovariable and multivariable logistic regression analysis. RESULTS: Monovariable analysis showed that there were significant differences in pediatric critical illness score (PCIS), procalcitonin (PCT) level, PaCO(2), blood lactate level, rates of intestinal or intra-abdominal infection, ascites, gastrointestinal dysfunction, mechanical ventilation, shock and multiple organ dysfunction syndrome (MODS) between the IAH and control groups (P < 0.05). Multivariable logistic regression analysis demonstrated that decreased PCIS, MODS, shock, gastrointestinal dysfunction and ascites were major risk factors for the occurrence of IAH. CONCLUSIONS: Children with sepsis who have decreased PCIS, MODS, shock, gastrointestinal dysfunction and ascites are at risk for the occurrence of IAH.

Transient ischemic attack induced by pulmonary arteriovenous fistula in a child: A case report.

BACKGROUND: Cerebral ischemic stroke is attributed to paradoxical cerebral embolism. Pulmonary arteriovenous fistula (PAVF) is a rare potential cause of cerebral ischemic stroke, and cerebral ischemic stroke induced by PAVF in children is rare. CASE SUMMARY: We report a case of right PAVF that presented as a transient ischemic attack (TIA) in a 13-year-old boy. The patient underwent embolization therapy and remained clinically stable for 2 years after treatment. CONCLUSION: TIA induced by PAVF in children is rare, lacks typical clinical manifestations, and should not be ignored.

[Correlations between serum interleukin-18 (IL-18) level, IL-18 gene promoter polymorphisms and the development of sepsis in children].

OBJECTIVE: To investigate the correlations of serum interleukin-18 (IL-18) level and IL-18 gene promoter polymorphisms to the development of sepsis in children. METHOD: Using enzyme-linked immunosorbent assay (ELISA), the authors tested the serum IL-18 level in 90 patients with sepsis and 123 normal controls, and their single nucleotide polymorphisms of the promoter region of IL-18 gene at position -607C/A and -137G/C were detected using polymerase chain reaction with sequence specific primers method and sequencing technique. RESULT: (1) The serum IL-18 level in sepsis groups was (196.56 +/- 157.32) pg/ml that was significantly higher than (66.16 +/- 41.63) pg/ml in normal controls (P < 0.01), the more severe the degree of sepsis was, the more significantly higher the serum IL-18 level was. The serum IL-18 level in non serious sepsis group was (152.87 +/- 114.96) pg/ml that was significantly higher than (66.16 +/- 41.63) pg/ml in normal controls, the serum IL-18 level in serious sepsis group was (191.98 +/- 169.72) pg/ml that was significantly higher than that in non serious sepsis group, and the serum IL-18 level in extremely serious sepsis patients was (323.89 +/- 159.35) pg/ml, the difference was highly significant (P = 0.000). The difference was significant among the groups with different severity of sepsis (P < 0.01). There was a negative correlation between PCIS (pediatric critical illness score) of sepsis and the serum IL-18 level (P < 0.01). (2) There were polymorphisms in IL-18 gene promoter of matched healthy children and sepsis in children. The GG genotype frequency (61.8%) of IL-18-137G/C in healthy children was the highest, followed by GC genotype (35.8%) and CC genotype (2.4%) in sequence. The G allele frequency (79.7%) was higher in IL-18-137G/C of healthy children than C allele (20.3%). The GG genotype frequency (71.1%) of IL-18-137G/C in septic children was the highest, the next were GC genotype (26.7%) and CC genotype (2.2%). The G allele frequency (84.4%) was higher in IL-18-137G/C of septic children than C allele (15.6%). The CA genotype frequency (61.0%) of IL-18-607C/A in healthy children was the highest, followed by CC genotype (26.8%) and AA genotype (12.2%). The C allele frequency (57.3%) was higher in IL-18-607C/A of healthy children than A allele (42.7%). The CA genotype frequency (76.7%) of IL-18-607C/A in septic children was the highest, followed by CC genotype (21.1%) and AA genotype (2.2%) in sequence. The C allele frequency (59.4%) was higher in IL-18-607C/A of septic children than A allele (40.6%). (3) The genotype frequency of IL-18-607 CA was 76.7% in sepsis groups that was significantly higher than 61.0% in normal controls, and the genotype frequency of -607 AA was 2.2% in sepsis groups that was significantly lower than 12.2% in normal controls, the difference was significant (P < 0.05). (4) In the order of -137CC, -137GC, -137GG, the serum IL-18 level in normal controls were as follows: (45.67 +/- 28.36) pg/ml, (53.27 +/- 37.91) pg/ml, (76.91 +/- 42.44) pg/ml, and with (140.50 +/- 60.10) pg/ml, (184.42 +/- 157.33) pg/ml, (237.02 +/- 161.76) pg/ml respectively in sepsis groups. In the order of -607AA, -607CA, -607CC, the serum IL-18 level in normal controls were: (48.80 +/- 32.11) pg/ml, (68.41 +/- 42.53) pg/ml, (70.17 +/- 43.87) pg/ml; and with (141.50 +/- 64.35) pg/ml, (151.21 +/- 121.19) pg/ml, (211.16 +/- 163.64) pg/ml respectively in sepsis groups. The difference was not significant among different groups (P > 0.05). CONCLUSION: The serum IL-18 level in sepsis groups was significantly higher than that in normal controls, which was related to the severity of sepsis. It was possible that the genotype of -607CA carriers was susceptible to sepsis, which mean that the genotype of -607CA might be susceptible genotype of sepsis. However, the genotype of -607AA might play an oppose role in the risk of sepsis.